Open 12. Rainfall Prediction Using Machine Learning 3

ABSTRACT
This study seeks a distinctive and efficient machine learning system for the prediction of rainfall.
The study experimented with different parameters of the rainfall from Erbil, Nicosia and
Famagusta in order to assess the efficiency and durability of the model. The neuro-fuzzy and
neural networks model is focused on this study. The learning of data is completed using hybrid
and backpropagation network algorithm. The rainfall parameters in this study are collected,
trained and tested to achieve the sustainable results through ANFIS and ANN models. The
monthly rainfall predictions obtained after training and testing are then compared with actual
data to ensure the accuracy of the model. The results of this study outline that the model is
successful in predicting the monthly rainfall data with the particular parameters. The training and
testing of data through neuro-fuzzy model helped in not only minimizing the errors up to RMSE
of 0.011, 0.015 and 0.025, but also maximizing the reliability and durability of the predicted
data. The results of the study highlight that the ANFIS model is most suitable among the
artificial networks for the rainfall prediction. The outcome data with ANFIS system presented
maximum accuracy with minimum error through the comparison between the actual data and
predicted outcome data.
Keywords: machine learning; neuro-fuzzy; neural networks; parameters; rainfall prediction
iii
TABLE OF CONTENTS
ACKNOWLEDGEMENTS ......................................................................................................... ii
ABSTRACT .................................................................................................................................. iii
ÖZET ............................................................................................................................................ iv
TABLE OF CONTENTS ............................................................................................................. v
LIST OF FIGURES ................................................................................................................... viii
LIST OF TABLES ....................................................................................................................... xi
LIST OF ABBREVIATIONS .................................................................................................... xii
CHAPTER 1: INTRODUCTION ................................................................................................ 1
1.1 Aim of the Study ................................................................................................................... 3

1.2 Significance of Study ............................................................................................................ 3
1.3 Limitations of the Study ........................................................................................................ 4
1.4 Problem Statement ................................................................................................................ 4
1.5 Methodology ......................................................................................................................... 4
1.6 The Study region and data ..................................................................................................... 4
1.7 Overview of the Study........................................................................................................... 5
CHAPTER 2: LITERATURE REVIEW ................................................................................... 6
CHAPTER 3: RAINFALL ......................................................................................................... 10
3.1 Introduction to rainfall ........................................................................................................ 10

3.2 Types of rain........................................................................................................................ 12
3.2.1 Conventional precipitation ............................................................................................... 12
3.2.2 Orographic rainfall ........................................................................................................... 13
3.2.3 Cyclonic or frontal rainfall ............................................................................................... 14
v
3.3 Measurement of rainfall ...................................................................................................... 15
3.3.1 Ordinary rain Gauge ......................................................................................................... 15
3.3.2 Self-recording rain Gauge ................................................................................................ 15
3.3.3 Zonal distribution of rain.................................................................................................. 16
3.4 Regime of rainfall................................................................................................................ 17
3.4.1 Equatorial rainfall regime................................................................................................. 17
3.4.2 Tropical rainfall regime .................................................................................................... 17
3.4.3 Monsoon rainfall regime .................................................................................................. 18
3.4.4 Mediterranean rainfall regime .......................................................................................... 18
3.4.5 Continental rainfall regime............................................................................................... 19
3.4.6 Maritime rainfall regime .................................................................................................. 19
CHAPTER 4: MACHINE LEARNING TECHNOLOGIES .................................................. 20
4.1 Introduction to machine learning ........................................................................................ 20

4.1.1 Artificial neural networks................................................................................................. 20
4.1.1.2 Neurons ......................................................................................................................... 21
4.1.1.3 Structure of ANN .......................................................................................................... 21
4.1.1.4 Weights.......................................................................................................................... 22
4.1.1.5 Feedforward neural network ......................................................................................... 22
4.1.1.6 Backpropagation algorithm ........................................................................................... 23
4.1.1.7 Nonlinear autoregressive exogenous model (NARX) ................................................... 23
4.1.2. Adaptive Neuro-Fuzzy Inference System ....................................................................... 24
4.1.2.1 ANFIS architecture ....................................................................................................... 24
4.1.2.2 Hybrid learning algorithm ............................................................................................. 27
CHAPTER 5: SIMULATION ................................................................................................... 28
5.1 Data processing ................................................................................................................... 28

5.1.1 Data Pre-Processing for Erbil ........................................................................................... 31
5.1.2 Data Pre-Processing for Nicosia ...................................................................................... 36
5.1.3 Data Pre-Processing for Famagusta ................................................................................. 41
5.2 Flowchart for rainfall prediction ......................................................................................... 46
vi
5.3 Selection of the input and output data ................................................................................. 47
5.4 Feature Extraction ............................................................................................................... 48
5.5 Training, Testing and Validation ........................................................................................ 49
5.6 ANN .................................................................................................................................... 49
5.6.2 Applying backpropagation and NARX model for Erbil .................................................. 49
5.6.3 Applying backpropagation and NARX model for Nicosia .............................................. 53
5.6.4 Applying backpropagation and NARX model for Famagusta ......................................... 56
5.7 ANFIS ................................................................................................................................. 58
5.7.1 Applying ANFIS for Erbil................................................................................................ 59
5.7.2 Applying ANFIS for Nicosia ........................................................................................... 60
5.7.3 Applying ANFIS for Famagusta ...................................................................................... 61
CHAPTER 6: DISCUSSION AND RESULT........................................................................... 65
6.1 Comparing results ............................................................................................................... 65

6.2 Actual and predicted data for Erbil ..................................................................................... 66
6.3 Actual and predicted data for Nicosia ................................................................................. 67
6.4 Actual and predicted data for Famagusta ............................................................................ 69
CONCLUSION ........................................................................................................................... 71
REFERENCES ............................................................................................................................ 72
APPENDICES ............................................................................................................................. 79
APPENDIX A: DATABASE FOR ERBIL .................................................................................. 80
APPENDIX B: DATABASE FOR NICOSIA .............................................................................. 81
APPENDIX C: DATABASE FOR FAMAGUSTA ..................................................................... 83
vii
LIST OF FIGURES
Figure 3.1: Heavy and unstable clouds of conventional rainfall .................................................. 12
Figure 3.2: Wave cloud formation on Amsterdam Island in the far Southern Indian Ocean ...... 13
Figure 3.3: Cyclonic rainfall cloud formation ............................................................................. 14
Figure 4.1: Neuron scheme .......................................................................................................... 21
Figure 4.2: Structure of ANN ...................................................................................................... 22
Figure 4.3: ANFIS architecture ................................................................................................... 25
Figure 5.1: Map of Iraq; (a) Erbil (b) focus area of this study (north Iraq) ................................. 29
Figure 5.2: Map of Northern Cyprus; showing Nicosia and Famagusta ..................................... 30
Figure 5.3: Trends in the distribution of rainfall for Erbil ........................................................... 31
Figure 5.4: Monthly average temperature for Erbil ..................................................................... 32
Figure 5.5: Trends in humidity for Erbil ...................................................................................... 32
Figure 5.6: Average wind speed for Erbil .................................................................................... 33
Figure 5.7: Correlation between humidity and rainfall for Erbil ................................................. 33
Figure 5.8: Correlation between temperature and rainfall for Erbil. ........................................... 34
Figure 5.9: Correlation between wind direction and rainfall for Erbil ........................................ 34
Figure 5.10: Correlation between wind speed and rainfall for Erbil. .......................................... 35
Figure 5.11: Trends in distribution of rainfall for Nicosia .......................................................... 37
Figure 5.12: Average temperature for Nicosia............................................................................. 37
Figure 5.13: Trends in humidity for Nicosia................................................................................ 38
Figure 5.14: Trends in wind speed for Nicosia ............................................................................ 38
Figure 5.15: Trends in average air pressure for Nicosia .............................................................. 39
Figure 5.16: Trends in wind direction for Nicosia....................................................................... 39
viii
Figure 5.17: Correlation for rainfall with humidity and temperature .......................................... 40
Figure 5.18: Correlation for rainfall with wind direction and rainfall with wind speed .............. 40
Figure 5.19: Trends in distribution of rainfall in Famagusta ....................................................... 42
Figure 5.20: Average temperature for Famagusta ....................................................................... 42
Figure 5.21: Trends in humidity for Famagusta .......................................................................... 43
Figure 5.22: Trends in average wind speed for Famagusta ......................................................... 43
Figure 5.23: Average wind direction for Famagusta ................................................................... 44
Figure 5.24: Trends in average air pressure for Famagusta ......................................................... 44
Figure 5.25: Correlation for rainfall with humidity and temperature .......................................... 45
Figure 5.26: Correlation between rain and wind direction and rainfall with wind speed............ 45
Figure 5.27: Proposed back propagation network architecture for Erbil ..................................... 50
Figure 5.28: Snapshot of regression using NNTOOL.................................................................. 51
Figure 5.29: Proposed NARX network architecture for Erbil ..................................................... 52
Figure 5.30: Training state for NARX ......................................................................................... 52
Figure 5.31: Proposed back propagation network architecture for Nicosia ................................ 53
Figure 5.33: Proposed NARX network architecture for Nicosia ................................................. 54
Figure 5.35: Proposed back propagation network architecture for Famagusta............................ 56
Figure 5.37: Proposed NARX network architecture .................................................................... 57
Figure 5.39: Proposed ANFIS architecture for Erbil ................................................................... 59
Figure 5.40: ANFIS rule viewer for Erbil.................................................................................... 60
ix
Figure 5.41: Proposed ANFIS architecture for Nicosia ............................................................... 61
Figure 5.42: ANFIS Rule viewer for Nicosia .............................................................................. 61
Figure 5.43: Proposed ANFIS architecture for Famagusta .......................................................... 62
Figure 5.44: ANFIS Rule Viewer for Famagusta ........................................................................ 62
Figure 6.1: Comparing actual and predicted data for Erbil using BPNN .................................... 66
Figure 6.2: Comparing actual and predicted data for Erbil using NARX ................................... 66
Figure 6.3: Comparing actual and predicted data for Erbil using Hybrid ................................... 67
Figure 6.4: Comparing actual and predicted data for Nicosia using BPNN ................................ 67
Figure 6.5: Comparing actual and predicted data for Nicosia using NARX ............................... 68
Figure 6.6: Comparing actual and predicted data for Nicosia using Hybrid ............................... 68
Figure 6.7: Comparing actual and predicted data for Famagusta using BPNN ........................... 69
Figure 6.8: Comparing actual and predicted data for Famagusta using NARX .......................... 69
Figure 6.9: Comparing actual and predicted data for Famagusta using Hybrid .......................... 70
x
LIST OF TABLES
Table 5.1: Correlation between input and outputs using SPSS for Erbil ..................................... 36
Table 5.2: Correlation between input and output using SPSS for Nicosia .................................. 41
Table 5.3: Correlation between inputs and outputs using SPSS for Famagusta .......................... 46
Table 5.4: Training, testing and validation by BP for Erbil, Nicosia and Famagusta ................. 63
Table 5.5: Training, testing and validation by NARX for Erbil, Nicosia and Famagusta ........... 63
Table 5.6: Training, testing and validation by Hybrid for Erbil, Nicosia and Famagusta ........... 64
Table 6.1: Obtained RMSE after training and testing .................................................................. 65
xi
LIST OF ABBREVIATIONS
AI: Artificial intelligence
ANFIS: Adaptive neuro-fuzzy inference system
ANN: Artificial neural network
ARMA: Auto-regressive moving average
BPNN: Back-Propagation Neural Networks
CBPN: Cascade-forward back propagation neural network
DMSP: Defense Metrological Satellite Program
DTDNN: Distributed time delay neural network
GEP: Gene expression programming
GPCP: Global perception climatology project
MSE: Mean Square Error
NARX: Nonlinear autoregressive exogenous model
NWP: Numerical weather prediction
PR: Precipitation radar
RMSE: Root-mean-square error
SPSS: Statistical Package for the Social Sciences
SSMI: Special Sensor Microwave Imager
TRMM: Tropical Rainfall Measuring Mission
xii
CHAPTER 1
INTRODUCTION
Rainfall play important role in forming of fauna and flora of natural life. It is not just significant
for the human beings but also for animals, plants and all living things. It plays a significant role
in agriculture and farming and undoubtedly; water is one of the most natural resources on earth.
The changing climatic conditions and the increasing greenhouse emissions have made it difficult
for the human beings and the planet earth to experience the necessary amount of rainfall that is
required to satisfy the human needs and its uninterrupted use in everyday life. Therefore, it has
become significant to analyze the changing patterns of the rainfall and try to predict the rain not
just for the human needs but also to predict for natural disasters that could cause by the
unexpected heavy rainfalls. To be more specific and aware of the devastating climatic changing
and stay updated; predicting rainfall has been the focus of computer scientist and engineers.
This study is focusing on predicting rainfall using Neuro-Fuzzy and Artificial Neural Network.
The rainfall prediction will not just assist in analyzing the changing patterns of rainfall but it will
also help in organizing the precautionary measures in case of disaster and its management. The
rainfall prediction would also assist in planning the policies and strategies to deal with the
increasing global issue of ozone depletion. The changing patterns of rainfall are associated much
with the global warming; that is increasing of the earth’s temperature due to increased
Chlorofluorocarbons emitting from the refrigerators, air conditioners, deodorants and printers
etc. that are the significant part of everyday life. The increasing temperature is actually affecting
the climate considerably (Sivakumar, 2006). Similarly, the rainfall prediction and weather
updates not only help in managing the macro level problems like flood and agricultural issues
because of poor or extreme rainfall (Lima & Guedes, 2015). The rainfall prediction could also
contribute to the well-being and comfort of the people by keeping them informed by tracking the
rainfall patterns and predicting the rainfall by Neuro-Fuzzy and Artificial Neural Network. The
rainfall predictions help the people to deal with hot and humid weather. The technological
development in the modern world has expanded the space for innovation and revolution.
Although the issues concerned are probably associated with these technological advancements
1
but one needs to consider the range of possibilities and opportunities that this technological
evolution has opened to the human beings.
In addition, the inappropriate or poor rainfall prediction is also one of the reasons that are
problematic in the water reserve management. The precise and correct rainfall prediction can not
only contribute to the effective and efficient utilization of this natural resource but it can also
help in managing the projects and plans for power generation. For this purpose, it is very
important to design and operate on a system that would assist in accurate prediction and easy
access to the users. Artificial Neural Network for rainfall prediction is one of the most suitable
and reliable systems for the rainfall prediction that has already benefited the operators for rainfall
prediction (Shaikh & Sawlani, 2017). ANN has the ability to access input information and
process it for a useful output. ANN does not need a previous knowledge of the processing of
information that gives it an advantage over other data processing systems (Darji, Dabhi, &
Prajapati, 2015). The rainfall prediction will also integrate adaptive Neuro-Fuzzy with ANN for
an increased accuracy and enhanced quality of the predicted output. To analyze the performance
of these algorithms; co-relation coefficient will be a key indicator in this study. ANN is the most
competent and effective tool for prediction of rainfall that actually contributes to the most
accurate forecasting (MuttalebAlhashimi, 2014). The Neuro-Fuzzy is also one of the effective
algorithms used for data analysis for the classification. It assigns categories and allocates cases to
similar groups/categories. So, each time a data is analyzed; it assigns that data to the most
suitable or most similar category it belongs to. This helps in making the regression and allows
the user to make a prediction for the similar sets of data or information received each time (Li,
Kwon, Sun, Lall, & Kao, 2009).
However, rainfall prediction with ANN using backpropagation and hidden layer approach
integrated with Neuro-Fuzzy is intended to produce precise and more accurate forecasts. The
predictions could be utilized for a maximum range of purposes and thus can play a vital role in
minimizing the issues associated with water reserves, agricultural problems with changing
climatic conditions and flood management. The appropriate utility and implication of the
estimated outcomes could also support the policy and development of strategies about resource
management and control with a variety of techniques and approaches that will actually impact
the human life in many ways.
2
1.1 Aim of the Study
The aim of the study is the prediction of the rainfall using historical monthly data based on
artificial intelligence methodologies such as Neuro-Fuzzy and artificial neural network. The
extraction procedures/algorithms will produce the output by classification of the data according
to the categories using Neuro-Fuzzy. The similar data will be grouped for the accurate and
precise information that will predict rainfall more correctly and with perfect figures. The
accurate and exact predictions will help in developing the more appropriate strategies for
agriculture and water reserves and will also be informed about the flood to implement
precautionary measures. The data for the rainfall prediction is collected from Metrology
Department of Erbil, Nicosia and Famagusta. This is the monthly data with all parameters of
rainfall including wind speed, direction, air pressure, humidity and temperature. The aim of the
proposed study is too effective and efficient in predicting the rainfall with accuracy and
precision.
1.2 Significance of Study
Rainfall prediction is significant not only on the micro but also on the macro level. The study is
of significance with respect to its vital contribution in the field of agriculture, water reserve
management, flood prediction and management with an intention to ease the people by keeping
them updated with the weather and rainfall prediction. It is also important to be utilized by the
agricultural industries for keeping their crops safe and ensure the production of seasonal fruits
and vegetables by updated rainfall prediction. The study will also be significant for the flood
management authorities as more precise and accurate prediction for heavy monsoon rains will
keep the authorities alert and focused for an upcoming event that of which the destruction could
be minimized by taking precautionary measures. The rainfall prediction will impressively help in
dealing with the increasing issue of water resource management; as water is a scarce resource
and it needs to get saved for the benefit of human beings themselves. Also, it will help the people
to manage and plan their social activities accordingly.
3
1.3 Limitations of the Study
i. The data sample is limited to monthly statistics only and does not provide the daily
output predictions.
ii. The climatic change and the global warming effect may impact the accuracy of the
expected output
iii. The locations for the data processing used in this study are geographically different and
distanced that could also impact the correlation efficient that will measure the
performance of the ANFIS and ANN in this research.
iv. The system discussed in this particular study will operate with Matlab software (R2017).
1.4 Problem Statement
The accurate and precise rainfall prediction is still lacking which could assist in diverse fields
like agriculture, water reservation and flood prediction. The issue is to formulate the calculations
for the rainfall prediction that would be based on the previous findings and similarities and will
give the output predictions that are reliable and appropriate. The imprecise and inaccurate
predictions are not only the waste of time but also the loss of resources and lead to inefficient
management of crisis like poor agriculture, poor water reserves and poor management of floods.
Therefore, the need is not to formulate only the rainfall predicting system but also a system that
is more accurate and precise as compared to the existing rainfall predictors.
1.5 Methodology
Spiral model of programming by creating a V0 and test it for feedback from the test sample. It
will retrieve the possible alterations to create the next Version of the algorithm. The test will be
by stimulating the neural network to retrieve results by archiving.
1.6 The Study region and data
The metrological data including humidity, air pressure, wind speed, wind direction and
temperature will be analyzed for three cities. Erbil in the North of Iraq and it has tall mountains
and experiences heavy rain every year. Although, it has very nominal humidity; the data
collected for Erbil has some appreciated predicted outcomes. Second is the capital of TRNC that
4
is Nicosia and the third city includes Famagusta that is a seashore city in TRNC. The data is
collected for 2012-2017 and it is collected on the monthly basis to predict the rainfall outcomes.
1.7 Overview of the Study
The thesis is designed as follows:
Chapter 1 is an introduction to the topic of the thesis. Chapter 1 outlines the overview of the
study, discusses the aims and significance of this study.
Chapter 2 reviewed the existing literature and highlighted the previous research on the proposed
thesis.
Chapter 3 is highlighting the general overview to the rainfall prediction and presents the
explanation of the rainfall prediction in the field of agriculture, water reservation and flood
prediction. It also grants information for the methods and approaches for the accurate prediction
in depth.
Chapter 4 is focused on the explanation of the artificial neural system with the ANFIS and
several modelling techniques like NARX algorithms are discussed in details.
Chapter 5 is an explanation of the stimulation of the data and presents the pre-processing and
correlation between the input and output. The application of ANN and ANFIS is also discussed
in detail.
Chapter 6 is the demonstration of results and discussions of the study. It also discusses root-
mean-square error to evaluate the best findings for analyzing the accurate rainfall prediction.
5
CHAPTER 2
LITERATURE REVIEW
Rainfall prediction is not an easy job especially when expecting the accurate and precise digits
for predicting the rain. The rainfall prediction is commonly used to protect the agriculture and
production of seasonal fruits and vegetables and to sustain their production and quality in
relation to the amount of rain required by them (Lima & Guedes, 2015).The rainfall prediction
uses several networks and algorithms and obtains the data to be given to the agriculture and
production departments. The rainfall prediction is necessary and mandatory especially in the
areas where there is heavy rainfall and it’s more often expected (Amoo & Dzwairo, 2016). There
are huge economies like those of Asia like India and China that that earn a large proportion of
their revenue from agriculture and for these economies; rainfall prediction is actually very
important (Darji, Dabhi, & Prajapati, 2015).
The rainfall forecasting is prevailing as a popular research in the scientific areas in the modern
world of technology and innovation; as it has a huge impact on just the human life but the
economies and the living beings as a whole. Rainfall prediction with several Neural Networks
has been analyzed previously and the researchers are still trying hard to achieve the more perfect
and accurate results in the field of rainfall prediction (Biswas, et al., 2016). The prediction of
seasonal rainfall on monthly basis by using the surface data to form annual prediction is also
essential for the agricultural activities and therefore the production and supervision of the
agriculture and crops. It could be done by recognizing the variations in the supply of moisture in
the air. The case of African region illustrates that how this succeeded and how West Africa
advantaged from the rainfall prediction in managing their agricultural activities (Omotosho,
Balogun, & Ogunjobi, 2000).
Similarly, the short-term streamflow forecasting for the rainfall is also reliable and bias-free. But
they are not much effective in predicting the flood and post-processing of rainfall prediction. An
approach called raw numerical weather prediction (NWP) was introduced in 2013, where the
approach focused on the Bayesian joint probability model to formulate prediction data. The
approach formed forecast possibility distributions for each location and it had prediction time for
6
it; collaborative forecasts correlated with space and time was produced in the Southern part of
Australia (Khan, Sharma, Mehrotra, Schepen, & Wang, 2015). This approach focused on Schake
shuffle to produce the forecast by the forecast possibility distributions (Robertson, Shrestha, &
Wang, 2013).
Furthermore, the short-term streamflow forecasting could also be used through the artificial
neural networks as researched by Zealand, Burn and Simonovic in 1999. The study conducted
outlined that ANNs ability to forecast for short-term stream flow and outlined some of the issues
that the approach encountered with ANNs (Kumarasiri & Sonnadara, 2006). Although, ANNs
with short-term stream flow can calculate and present complex and nonlinear relationship
between input and output with an ability to outline the interface effect as well but has issues in
processing some input data with certain type and number. The ANNs also encountered difficulty
with dimensions of the hidden layers. This research outcome was represented by the data of
Winnipeg River system in Ontario, Canada using the quarter monthly data. The outcomes of the
study were encouraging with AANs performed quite well for the four prediction lead-times. The
RMSE for the test data of 8 years outlined variation from 5cms to 12.1cms in a forecast from
four-time step to two-time step ahead respectively (Zealand, Burn, & Simonovic, 1999).
Also, the recent decade highlighted the significance of artificial intelligence and it has gained
attention in water resource management and engineering as well. ANNs, ANFIS and GP are the
driving simulations of AI and they are advantaged over other systems and approaches because of
being more reliable and competitive. The adaptive neuro-fuzzy inference system (ANFIS) for
time series and ANN for predicting streamflow in Apalachicola River, the United States with
that of other neural network techniques like hybrid (Mittal, Chowdhury, Roy, Bhatia, &
Srivastav, 2012); when compared to wavelet-gene expression’ programming approach outlined
the following results; ARMA model predicting accurate results for 1 day ahead time whereas,
ANFIS forecasted the results for 2 days ahead time. The results from AI using ANFIS were more
accurate and could predict 2 days ahead of time data rather than GEP and ANN (Nayak,
Mahapatra, & Mishra, 2013). But for the 3 days forward data; ANN performed better than other
models. For the monthly data; ANN, ANFIS and GEP outperformed as compared to ARMA
models in the first part of the study (Karimi, Shiri, Kisi, & Shiri, 2016).
7
Water as is one of the most useful resources of the earth. There is no human and living thing on
earth that can survive without water. As, this precious resource is running out because of the
increasing temperature of the earth and the unexpected and unappreciated climatic conditions
due to global warming. (Mittal, Chowdhury, Roy, Bhatia, & Srivastav, 2012). In addition, the
comparison among different neural models revealed that Non-linear autoregressive exogenous
networks (NARX) and back propagation neural BPN) performed better than distributed time
delay neural network (DTDNN) cascade-forward back propagation neural network (CBPN) in
outlining more accurate and precise results for rainfall prediction (Devi, Arulmozhivarman,
Venkatesh, & Agarwal, 2016). In comparison, statistical forecasting methodology can also be
used for the rainfall prediction that outlines by using two different approaches like traditional
linear regression and polynomial-based nonparametric; where nonparametric method outlined
more competing results. Both the approaches could predict the 1-3 monthly rainfall forecasting
data that could actually impact water resource planning and controlling (Singhrattna,
Rajagopalan, Clark, & Kumar, 2005). The periodic and episodic rainfall data for the south-west
peninsula of England has also exposed that atmospheric characteristics are key players of
outlining the monthly and seasonal forecast (Mcgregor & Phillips, 2003).
The rainfall prediction is also emphasized for its significance for the prediction of flood and
consequently takes the precautionary measure to save the people from devastating destructions
that a flood can cause (Hoai, Udo, & Mano, 2011). There are studies that outlined the
significance of rainfall prediction in forecasting flood on the regions where there is heavy rain
every year. The areas with high risk for flood are the vulnerable areas that need the rainfall
forecasting not just to save a human life but to safe agriculture, water reservation and livestock
(Fang & Zhongda, 2015).
In comparison, the significance of rainfall prediction is also important for areas with high
probability for the drought. The areas with high drought seasons are also vulnerable to high risk
in terms of agriculture and livestock with an extreme threat to human life as a whole; the study
conducted for Sakae River basin of Thailand (Wichitarapongsakun, Sarin, Klomjek, &
Chuenchooklin, 2016). The artificial neural network model for rainfall prediction of 1to 6 hour
ahead time is studied for Bangkok, Thailand by Hung, Babel, Weesakul, and Tripathi in 2008.
The study outlined that within artificial neural networks, using six models utilizing rainfall
8
parameters like humidity, air pressure, wind direction and wind speed can give more accurate
and precise prediction when previous forecasting data is also used with these parameters as an
input as well (Hung, Babel, Weesakul, & Tripathi, 2009).
Nevertheless, land sliding is another natural hazard that could be caused due to heavy rainfall.
The rainfall prediction could assist in combating the devastation caused by land sliding. The
rainfall prediction for the areas vulnerable to land sliding is an essential part of artificial
intelligence within engineering and management fields (Schmidt, Turek, Clark, Uddstrom, &
Dymond, 2008). The metrological and hydrological centres are struggling hard to produce the
more competitive and precise rainfall prediction in order to overcome these issues that the
rainfall can cause and their efforts have marked quite an improvement in the rainfall prediction
and forecasting data for many models using the neural networks. The prediction for extreme
rainfalls is useful for not just the metrological departments in sharing in time alerts but also for
the hydrological departments in order to form better safety measures for example the flood
prediction in Australia (White, Franks, & McEvy, 2015).
The rainfall prediction systems are much popular with artificial neural networks and the rainfall
prediction departments like the metrology and hydrology engineering with management
(Abhishek, Kumar, Ranjan, & Kumar, 2012). The rainfall prediction using the neural network
aims at predicting more efficient and more accurate results and precise predictions for a more
useful and reliable output that could be used by the management and engineering departments in
designing the plans and policies that will not only increase efficiency but it will also enhance the
management systems from a quality data produced by using the Artificial Neural Networks. The
study conducted with the different networks highlighted different results by operating within
same training functions and outlined that back propagation neural network is capable of
obtaining more precise predictions. Also, that increased neurons can decrease errors (MSE)
(Sharma & Nijhawan, 2015). Neural networks have proved capability for the rainfall prediction
and in obtaining accuracy with precision among the other networks with other modelling
techniques (Narvekar & Fargose, 2015).
9
CHAPTER 3
RAINFALL
3.1 Introduction to rainfall
Rainfall is one the most significant atmospheric occurrence that is not only useful for the
environment itself but for all the living beings on the earth. It affects everything directly or
indirectly and because it is one of the most important natural phenomena; it is also important for
the human beings to ponder on the precipitation changes with the change in climate (Alpers &
Melsheimer, 2004). The rainfall has a significant impact on the universal gauge of atmospheric
circulation and it affects the local weather conditions as well. The rainfall helps in balancing the
increasing temperature and in the survival of the human beings (Trenberth, 2011). The increasing
temperature of the world is associated with the global warming and the water is one of the scarce
and most useful resources which in the result of this increasing temperature are evaporating from
the reserves. Rainfall is also compensation to all these reserves and it is necessary for the
agriculture and its production as well. The phenomenon of rainfall differs with the difference in
latitude and longitude. The rainfall phenomenon also differs with the difference of regions,
planes, mountainous and plateaus (Alpers & Melsheimer, 2004).
Rainfall occurs as stratiform or convective rain; the high latitude areas experience stratiform rain
which is quite a dominant form of the rain. These areas include the tropical and subtropical and
they experience 50% to 80% of stratiform rain precipitation (Alpers & Melsheimer, 2004). It is
important to measure the distribution of the rainfall on the global level and for that currently the
remote satellite sensing techniques are assisting in measuring the distribution of the rain on the
global level. Special Sensor Microwave Imager (SSM/I) onboard with the US Defense
Metrological Satellite Program (DMSP) are used for gathering the information about the rainfall
with other space-borne instruments like microwave instrument , flying aboard the US –Japanese
Tropical Rainfall Measuring Mission (TRMM) and precipitation radar (PR) that operate on
different frequencies and are assisting in the data collection and in getting the footprints
accurately (Alpers & Melsheimer, 2004).
However, the precipitation of the rainfall is not constant and it changes every year. The rainfall is
actually the evaporated water from the earth surface because of high temperature or heat that
10
goes up and then comes down in the form of rain or snow (Alpers & Melsheimer, 2004). The
rainfall is the most significant phenomenon has always been associated with the increasing
demands of mankind. The human beings on the earth cannot live without water and there is no
way that they can also prude it artificially. It is one of those precious resources that cannot be
artificially produced and thus it is the focus of the most studies and researches going on in the
world. the scientists and the engineers are collaborating researchers to find out the best and most
effective way of measuring the rainfall and predicting the rainfall to compensate for the extreme
water use around the globe and to be sufficient for the increasing demands in terms of
agriculture, water reserves and in order to be safe and sound from the natural disasters like the
flood and land sliding. There is a need to focus on the efficient use of the water and to make the
accurate predictions about the rainfall so that the time and the resource could be saved
(Trenberth, 2011).
Similarly, the frequency, intensity and the amount of precipitation are changing with the
changing temperatures and the effect of heat on the environment is also causing the changes in
the precipitation levels (Kumar, Yang, Goddard, & Schubert, 2004). The rainfall can vary from
the tropical storms to a thunderstorm, orographic rainfall and cyclones. The changing
precipitation levels are observed by the Global Precipitation Climatology Project (GPCP) and
presented the global changes in precipitation by the changing lands and the time period to impact
it (Gu, Adler, Huffman, & Curtis, 2007). The rainfall affects the surface gravity waves in the
upper water layer by generating turbulence and enhancing the roughness on the sea surface
(Nystuen, 1990). The rainfall causes the notable changes to the environment and does not only
assist in the everyday demands but also in cleaning the environment for the human beings. The
air after the rainfall is fresh and clean and the pollution caused by the human beings is also
controlled by the rainfall. The distribution and the amount of rainfall differ from region to region
and area to area; therefore, with some precipitation in the different area it can also cause floods,
tsunamis and land sliding. Therefore; the prediction of the rainfall and the forecasting of the
precipitation is quite a significant field on which the scientists and the researchers are exploring
ways to accommodate this natural phenomenon and to manage the resource in a more
appropriate and useful manner to multiply the human life.
11
3.2 Types of rain
3.2.1 Conventional precipitation
The significant and the most dominant form of the rainfall is the convectional rainfall. It is
experienced in the high latitude areas like the tropical and the subtropical. It is usually observed
with lightning and thunderstorm. The conventional rainfall is a type of rainfall that is affected by
the mountains and the mountainous regions; as it is the most dominant form of rainfall and it
depends on the latitude (Collier, 2003). The formation of the conventional rainfall occurs when
the air on the surface of the earth gets intense by the heat of the sun. The hot air is lighter than
the cool air so it evaporates from the earth surface and forms clouds in the atmosphere. The
further rise in the water vapours and gradually these vapours move upward direction towards the
area of converging air and forms thick and heavy clouds. The heavy and unstable clouds rise
further and the instability of these clouds then compel them to fall on the surface of the earth
again in the form of conventional rain (Selase, Agyimpomaa, Selasi, & Hakii, Precipitation and
Rainfall Types with Their Characteristic Features, 2015).
The conventional form of rainfall is observed mostly in West Africa and it is always followed by
a thunderstorm and heavy lightening because of the heavy and unstable clouds rising upward in
the atmosphere with converging air (Blyth, Bennett, & Collier, 2015). The figure 2 below shows
the instability of the heavy clouds of convectional rainfall.
Figure 3.1: Heavy and unstable clouds of conventional rainfall (Collier, 2003).
12
3.2.2 Orographic rainfall
The orographic rain is the form of rainfall that is formed by the moist air which usually can be
observed above the mountains. The moist air above the mountains is evaporated or lifted upward
direction. When the moist air is lifted and rises to a certain level it cools down; the orographic
clouds are formed and then condenses and forms the precipitation. The orographic rainfall is
formed by the midlatitude lands like the one with large mountains (Gray & Seed, 2006). The
orographic rainfall has tiny water drops that are condensed. These small water drops from clouds
and then these small clouds come together to form bigger clouds. These clouds also turn into
snow over some period of time (Jr., 2012).
The orographic rainfall is observed on the midlatitude mountains with an axis perpendicular to
the prevalent wind direction. These directions cause the sharp rainfall transitions and could be
observed better with two adjacent ranges of the mountain to circulate the moist air more. The
steadier and these are experienced mainly in afternoon of the summers with dynamic
thunderstorms. The discrete formation of orographic precipitation is sometimes observed on the
small mountains as well (Roe, 2005). Orographic rainfall is due to the uplift of masses of air by
the wind (Smith & Evans, 2007).
Figure 3.2: Wave cloud formation on Amsterdam Island in the far Southern Indian Ocean
(NASA, 2005).
13
3.2.3 Cyclonic or frontal rainfall
The cyclonic or the frontal rainfall is the last and third type of the rainfall. The cyclonic by name
represents the tempesting and occurs when the air masses with distinct characteristics collide
with one another. The collision of light air that is warm and the cold air that is heavy occurs; the
cold air encourages the warm air because it is lighter to rise. The rising air cools down by
forming the water vapours. The condensation process initiates and forms the clouds (Thatcher,
Takayabu, Yokoyama, & Pu, 2012).
The formation of clouds become heavy as they meet with other clouds and these heavy clouds
become unstable and fall back on the earth as cyclonic rainfall. The cyclonic rainfall is common
in Tropical areas with 23% of Degrees North Latitude and South of the equator with the
temperate zone latitude of 66% degrees North and South. This is the reason that it is also known
as the frontal rainfall (Thatcher, Takayabu, Yokoyama, & Pu, 2012).
The frontal/cyclonic rainfall has a specific period and time when it is more dominant and the
precipitation is more rapid and concentrated. The cyclonic/frontal rain may have an extended
precipitation that could be extended and keep the weather wet for log days. It is a lognormal type
of distribution and its distribution depends on the area of precipitation (Cheng & Qi, 2001).
Figure 3.3: Cyclonic rainfall cloud formation (Rodgers & Adler, 1980)
14
The above figure 3 represents the formation of cyclone rainfall or the frontal rainfall in the
tropics and highlights the changing data readout orbit that is increasing with the increase in the
cyclonic visibility as shown in the four pictures (Rodgers & Adler, 1980).
3.3 Measurement of rainfall
The rainfall is a natural phenomenon that is measured in mm. The measuring instrument is
203mm in diameter. This is a funnel that gathers the rainfall into a cylinder and has the capacity
of measuring u to 25mm of rainfall (Alpers & Melsheimer, 2004). There are two techniques for
measuring rainfall as described below:
3.3.1 Ordinary rain Gauge
The ordinary rain gauge measurement is a less effective and less accurate technique of measuring
the rainfall. It has been observed that the ordinary gauge is the non-automatic observation and
uses a glass to measure the rain at regular intervals. It has a shell, a storage bottle with a storage
vessel and a glass for measuring the rain. It is not effective for the heavier and substantial rainfall
like the cyclonic/frontal rainfall (Carvalho, Assad, Oliveira, & Pinto, 2014). It uses a rainfall
record book to compare and measure the rainfall for a particular period. It is less accurate and the
data collected may not be precise (Agnihotri & Panda, 2014). The ordinary method of the rainfall
measurement is helping in the local level measurements and those that are less accurate and less
precise but this method of rainfall collection is appropriate for the measurement t record for a
larger level. The observations performed in the ordinary rainfall are manual so the errors are not
minimized.
3.3.2 Self-recording rain Gauge
The conventional self-recording rain gauge is more efficient and more effective in measuring the
rain than that of the ordinary gauge for the measurement of rain. The traditional method of
recording rain is inefficient and gives inaccurate results as it’s done manually so it can involve
human errors (Beard, 1962). The self-recording rain gauge is observed to use simple technique
and instruments to produce better measurements in order to have better probability and accuracy.
The self-recording gauge consists of a tipping bucket and a lever balance that weighs the rain
15
accordingly. It has a marking pen which marls each certain level recorded with the movement of
the bucket and the precipitation for each hour is recorded by the gauge automatically without the
support of human beings (Hansen, 1961).
3.3.3 Zonal distribution of rain
The patterns of the rainfall precipitation are not constant; it varies season to season and location
to location. There are certain different zones that get more precipitation than few of them getting
less precipitation. The precipitation of the rainfall as a mean global distribution studied to be
affected by the latitudinal zones, land and sea surfaces and precipitation. The East Asian region
precipitation including China, Korea and Japan evaluates that monsoon starts from mid-end of
May to the end of July for China and September in case if Korea. It ends by the early August
Peninsula and for Japan; it lasts longer from mid-September to end of October. (Qian, Kang, &
Lee, 2002).
The Middle Eastern region experiences a severe autumn rainstorm. The countries along the red
sea are the most significant to experience this kind of rainfall distribution and the Mediterranean
countries are more unstable to these conditions. The experience intense thunderstorms but the
weather for summer is more of hot and dry. These countries also experience hailing and may
sometime encounter flooding. The North African region is also included within this area and
experience the same weather. Both the weathers are extreme; in case of summers and winter
being on the cold front (Dayan, Ziv, Margalit, Morin, & Sharon, 2001).
The rainfall distribution in the region of East Africa is also studied as significantly more than that
of the amount of rainfall studied. The patterns of zonal distribution are observed to be variable
for each season. The rainfall prediction for each day is also changing with the changing trends of
rainfall. The zonal distribution of the rainfall is open to seasonal change and the change in the
sea level and earth surface due to the movement of plates (Johnson, 1962). The parameter like
wind, atmosphere and humidity are also significant in determining these patterns. Air pressure
and temperature also affect the zonal distribution of the rainfall annually and monthly. The
changing temperature of the earth is also influencing the zonal distribution because it has more
chances of being deteriorated. The high precipitation of the rainfall distribution recorded is that
of an equatorial zone. Southeast Asia and the middle latitude areas experience the comparatively
16
low amount of rain distribution and the deserts of subtropical regions experience even fewer or
extremely insufficient rainfall annually (Adler, Huffman, Bolvin, Curtis, & Nelkin, 2000).
3.4 Regime of rainfall
According to Haurwitz and Austin in 1944; there are six main regimes of the rainfall that are
described as:
3.4.1 Equatorial rainfall regime
It is the form of rain that is characterized by experiencing the rainfall throughout the year and in
all seasons. This character of rainfall is not outlined by the rainfall in a particular season or in
few specific months rather it is characterized by the consistent and continuous rainfall annually.
The equatorial rainfall regime is experienced to have heavy rainfall in the month of March and
September. The thermal air currents are generated by the heating effect and that contributes to
the formation of heavy and unstable clouds resulting in the extreme rainfall within these months
of the year. The equatorial regime of the rainfall extends between the zone of 10⁰ N and 10⁰ S
latitude. The equatorial rainfall regime is followed by thunderstorm and heavy lighting because
of the instability and huge size of the clouds. The rainfall is observed to be in the form of heavy
showers but the time duration observed for these types of regimes is quite less as compared to
the other regimes. These are the low-pressure belts and relatively high temperatures. Mostly they
experience the conventional rainfall because of the heating effect. The distribution of the rainfall
throughout the year is equal and uniform. These regions experience heavy rainfall sometimes
with hailing and storms during the year (Haurwitz & Austin, 1944).
3.4.2 Tropical rainfall regime
The tropical rainfall regime is characterized by the heavy rainfall not throughout the year but
only in the summers. The winters of the tropical areas are usually dry and they are not associated
with much rainfall. The summer season not only experiences the heavy rainfall but is
comparatively pleasant because of the consistent rainfall. The northern hemisphere and the
Eastern hemisphere usually experience the maximum and minimum rainfall during the months of
July and December. The tropical regime of the rainfall is under the influence of summer
17
stagnation and thus it can be said that the winters are significantly dry. Therefore, the areas
occupying the tropical regime may suffer in the winters but are ready to shine in the summers
(Haurwitz & Austin, 1944).
3.4.3 Monsoon rainfall regime
The monsoon rainfall regime is characterized as the zone 4 that is also under the stagnation of
the summer season. The rainfall experienced in this regime is characterized as only in summers
but slightly less as compared to that of tropical. The winters are also extremely dry as that of the
tropical because this regime is at the sub-tropical high-pressure belt. The maximum rainfall is
experienced in the month of July that is known to be the monsoon season and continues to the
end of August. There are not much heavy showers of the rainfall and the thunderstorm and the
lightning is also merely observed but the light shower continues for almost 2 months day to day
and time to time. This regime is associated with the monsoon winds; as the monsoon winds start
to blow; like the phenomenon experienced on the orographic rainfall; the wind elevates the water
vapours and condenses them to form clouds and then into rainfall shower (Haurwitz & Austin,
1944). The winds observed to move the water vapours in the upward direction are also associated
with the formation of the cyclonic rainfall as described above. The air of different masses
collides with one another and initiates the condensation process. The lighter air that is warm air
collides with the heavy cold air to form the clouds and thunderstorms or lightening effects could
be observed for the monsoon regime of the rainfall.
3.4.4 Mediterranean rainfall regime
The Mediterranean rainfall regime experiences the driest weather. For example: the weather
conditions of Nicosia and Famagusta. The weather of these cities are extremely dry not only in
winters but also in summers. The regime is associated with the sub-tropical high-pressure belt for
the whole year. Although, the winters are cold and rainy they last for only 2 months that of
January and February and the summer season is extremely hot and dry. They experience the
comparatively heavy rainfall in the winter season opposite to the above-mentioned regime sand
but dryness prevails throughout. The summers despite being very hot for the Mediterranean are
still dry and hardly the rainfall is experienced. The city of Nicosia and Famagusta are also
following this belt ad experience exactly the same weather as observed from the data collected
18
from the metrological department. The Mediterranean rainfall regime is defined as zone 4 by
Haurwitz and Austin, as extremely dry (Haurwitz & Austin, 1944).
3.4.5 Continental rainfall regime
Similarly, the continental rainfall regime experiences the heavy rainfall in summers due to the
convection. The high temperature forces the water to evaporate in the form of water vapors from
the water surface and form clouds. The summer season continuously experiences the convection
and the rainfall is experienced throughout the summer season opposite to that of the winter
season formation of the clouds. Because the summer season experiences the heavy rainfall
therefore, the winter season is usually dry it experiences only few slight showers of the
precipitation. The weather conditions for such a regime is neither very dry nor are they rainy.
The weather is moderate and the extreme seasons are not experienced. The high precipitation
within the continental regime could experience the cyclonic rainfall and conventional rainfall as
the precipitation depends upon the temperature. The high temperature will impact the
precipitation and the formation of the clouds for heavy showers of the rainfall (Haurwitz &
Austin, 1944).
3.4.6 Maritime rainfall regime
The maritime regime is characterized by the mid-latitudes and along the western margins of the
continent. The maritime regimes experience the rainfall in winter season and the summer is not
very dry but due to the high precipitation in the winter season; the maritime regimes experience
usually the dry summers. These regimes also experience the cyclonic rainfalls with several
intervals and the process of rainfall continues throughout the year. There is no season in these
regimes that is extremely dry or experiences no rainfall. These regimes also experience the
winter monsoons despite of the summer monsoon winds and are expected to observe with the
monsoon showers in the winters rather than that of the summers. The maritime air is conveyed to
the cost through the dominant westerly (Haurwitz & Austin, 1944).
19
CHAPTER 4
MACHINE LEARNING TECHNOLOGIES
4.1 Introduction to machine learning
Machine learning is associated with the study of the algorithms that enhance the efficiency of the
machines/computers automatically through the training and testing of the machine/computers
with certainly different variables. The machine learning is among the most favourable and fastest
growing areas of computer technology. The computers work efficiently with different algorithms
and functions. The machine learning is the training the computer with certainly different
algorithms to experience the machine in automatic smart data processing. The machine learning
enhances the efficiency and accuracy of the data processing and is used in a wide range of fields.
The machine learning is developed with effectual algorithms that utilize a certain set of tools and
functions to solve the complex and huge data. The machine learning is assisting in the diverse
field; these are normally artificial intelligence applications that are used for recognition and
prediction like that of computer engineering and medical fields. The machine learning is popular
among the modern computer technology and has many benefits. Machine learning develops
some rules for the input data as discussed in the hybrid model that helps the machine to process
the similar case each tie efficiently. It works on the prediction, and it is more important to
understand that how variables the inputs into are moved into vectors. The machine learning has
minimized the manual jobs for the people that also could have the space for errors and
inaccuracy (Smola & Vishwanathan, 2008)
4.1.1 Artificial neural networks
The artificial neural network is a computer networking system that can perform huge and
intelligent tasks. It is a parallel and distributed processing system that can accomplish most
complex tasks of recognition, prediction and detection without increasing the complexity of the
problem. The artificial neural network has one input layer and one output layer between which
are the hidden layers that process data. Each layer by processing the data forwards the result to
the next hidden layer and finally the output layer obtains the result after the data processing. The
artificial neural network is one the most popular machines of artificial intelligence that are used
20
almost in every field nowadays. It uses certain different models for processing data like feed-
forward back propagation, NARX model with different functions for each model. These are
dynamic machines capable of solving complex to everyday problems and made the human life
easy (Yegnanarayana, 2009).
4.1.1.2 Neurons
The artificial neural network performs these complex and huge tasks through the neurons that are
fed into the layers and process the data just like the human brain. Neurons operate in the human
brain to perform all the tasks and so are the artificial neural network does. The neural network is
non-linear functions. The neurons in the neural network are first trained with the old data in
order to get the new and predicted data.
Figure 4.1: Neuron scheme (Skorpil & Stastny, 2006)
After the training, testing is carried out to check the different results with different data and to
obtain the comparison by feeding the system with a different number of neurons. The number of
neurons fed to the system is varying and depends on the data and processing complexity.
Therefore, the architectures may differ from one another depending on input/output complexity
and the layers in the system (Demuth, Beale, Jess, & Hagan, 2014).
4.1.1.3 Structure of ANN
The architecture of the ANN has three layers with a large number of neurons, the neurons are
called the units and they are arranged in a sequence of layers.
21
Figure 4.2: Structure of ANN (Ahn, 2017)
INPUT LAYER is the first layer of the ANN structure is the input layer that takes the input for
processing.
HIDDEN LAYER is the second layer that process the data transferred from the input layer for
processing through neurons and, the weights are updated continuously for precision and validity
of the output
OUTPUT LAYER is the third and the last layer through which the results are obtained from the,
as shown in figure 4.2 above.
4.1.1.4 Weights
The weights in the ANN architecture are the memory storage that stores the information and data
to get the desired results. The weight during the training, testing and validation are modified at
every step so that the accuracy of the output is achieved; also, they store data for the future
operations.
4.1.1.5 Feedforward neural network
The feed-forward neural network has multilayers for the processing of elements. Each layer
processes the input data that it receives and forward the results obtained to the next layer. For
this processing, each layer operates independently to generate the resulting that is forwarded to
the next layer. The result obtained through processing of each layer is ultimately obtained from
the output layer. Between input and output layer; there are hidden layers. The elements that
process the input data work like the neurons in the human brain, these are called artificial
neurons. The neurons in the layers send messages or information to other neurons through a
channel called connections.
22
4.1.1.6 Backpropagation algorithm
The feed forward back propagation is used to detect the error and consequently highlight the
performance of the network using the certain inputs, number of neurons and to check the validity
and accuracy of the output obtained. In the back-propagation model by the ANN; weights are
decrypted and adjusted in the neural network. The system performs several cycles of
backpropagation with the input data to get the desired output (Y.H.Zweiri, J.F.Whidborne, &
L.D.Seneviratne, 2002). The backpropagation a very simple yet efficient algorithm, it consists of
N processing elements with functions of input and output as below.
y = G(x, W) (4.1)
In this equation, x is the input vector, y is the output vector and W is the propagation error
weight matrix, the later matrix is shown by equation 4.2.
W = (𝒘𝟏 T, 𝒘𝟐 T,...., 𝒘𝑻𝑵 )T (4.2)
In equation 2,𝑤1,𝑤2 , ..., 𝑤𝑁 are the individual vectors given by in equation 4.3 below.
(4.3)
4.1.1.7 Nonlinear autoregressive exogenous model (NARX)
The NARX is the model that is found from Autoregressive with Exogenous Input model. The
NARX is a nonlinear and recurrent dynamic model. It is a feedback neural network which is
efficient in obtaining the output results that are accurate and precise. It is suitable for modelling
nonlinear systems like the artificial neural network. The NARX is best for learning with gradient
23
algorithm. The gradient descent is obtained accurately with NARX. The below equation 4.4
illustrates the algebraic expression of NARX.
(4.4)
The NARX equation in the vector form can be written as below equation 4.5.
(4.5)
The NARX models are used popularly for the identification and recognition tasks. The
predictions are and forecasts could also be made efficiently by using the NARX models. They
operate under certain different functions and are autoregressive. This model uses feedback
connections (that are neurons sending the information to other neurons) in the several layers of
the network for enhanced accuracy. It is based on ARX model that is linear and used to predict
the time series commonly (Khamis, Nabilah, & Abdullah, 2014).
4.1.2. Adaptive Neuro-Fuzzy Inference System
The ANFIs is an efficient machine learning and artificial intelligence network that is sometimes
advantaged over the neural networks. The ANFIS aims at reducing the complexity of the
operation and simplify it to get the desired results and output. It also uses the neurons for
processing the data, the neurons work as nodes. The neuro-fuzzy system introduces a set of rules
for each operation that also stores the data and information for the future operations. The rules
introduced depend on the inputs and outputs. It has a domain knowledge which is commonly
practised for obtaining the outputs. The concepts of adaptive networking are used with certain
techniques to process the desired output. The output depends on the updating parameters and
their collection. The node is the processing unit of the neuro-fuzzy. The ANFIS drives the rule
for the different optimization techniques (Wahyuni, Mahmudy, & Iriany, 2017).
4.1.2.1 ANFIS architecture
The below figure 4.3 illustrate the ANFIS architecture for two inputs with five layers. The
ANFIS makes rules while processing the data that are used for accurate and precise results.
24
These rules also help to operate the future data for maximum efficiency (Wahyuni, Mahmudy, &
Iriany, 2017). The architecture has five layers as described below:
Figure 4.3: ANFIS architecture
LAYER 1 (MEMBERSHIP FUNCTION)
The first layer has nodes and each node i in this layer is an adaptive node to node to function, as
outlined in the equation 4.6 below.
(4.6)
In the above equation 1, x and y are the inputs at node i, Ai and Bi are the linguistics label. The
O1𝑖 and O1𝑖 -2 are the membership functions of Ai and Bi. The member ship function is based on
the linear equation curve that specifies the maximum value as 1 and minimum value as 0. The
membership function A has parameters as shown in equation 4.7 below.
(4.7)
25
LAYER 2 (RULES LAYER)
The layer 2 in the architecture has fixed node. The output obtained through this layer is the
product of all the inputs that are fed into this layer. To calculate the output through this layer,
following equation 4.8 can be used.
(4.8)
LAYER 3 (NORMALIZED FIRING STRENGTH)
Similar to the layer 2, each node in this layer is also fixed. In this layer, normalized firing
strength is the ratio of the output i on the preceding layer to the whole output of the preceding
layer. It is represented in equation 4.9 below.
(4.9)
LAYER 4 (DEFUZZIFICATION)
In this layer, i is a node to node adaptive function. The weight obtained from layer 3 and the
parameters to two inputs are obtained by linear regression function of order 1, as shown in
equation 4.10 below.
(4.10)
LAYER 5 (ADDITION)
The last layer 5 gathers all the inputs obtained from each layer and adds them up for the final
result. The sum function is outlined in the equation 4.11 below.
(4.11)
26
4.1.2.2 Hybrid learning algorithm
In the ANFIS architecture there are five layers. The first and the fourth layer contain the
parameters that can be updated time to time. But the first layer is nonlinear while the fourth is
linear. Therefore, both the parameters need to be updated through the learning method which is
capable of training linear and nonlinear simultaneously, hybrid system is the one introduced in
1993 by Jang, that train both layers at the same time (Faulina & Suhartono, 2013). The ANFIS
network is trained through Hybrid learning algorithm. It uses descent gradient to denote the
errors by forward pass and backward pass in order to train layer 1 and layer 4 at the same time.
The error can be measured by the equation 4.12 below.
(4.12)
Tm,p is the mth component of the pth target
OLm,p is the mth component the actual output vector
The overall error is as follows,
(4.13)
27
CHAPTER 5
SIMULATION
5.1 Data processing
In this study, we are considering the prediction of rainfall. The main purpose and aim of the
study are to utilize the machine learning for predicting the rainfall with accuracy and precision.
Therefore, the study includes and examines the source data in the pre-processing phase and
utilizes this data in the stimulation or processing phase further to predict the output that is
forecasting effectively and efficiently. For this purpose, statistical data of Erbil, Nicosia and
Famagusta is collected from the metrological departments. The data collected from the
metrological departments is monthly, as this study will also focus on the monthly rainfall
prediction. The data includes the significant parameters of the rainfall. Temperature, wind
direction, wind speed, air pressure and humidity are considered. Each parameter of the rainfall is
analyzed separately for Erbil, Nicosia and Famagusta. The statistical data studied and examined
in this study will be used as input for processing the output as rainfall prediction. The data
processing is done through machine learning techniques and different neural networks like
ANFIS and NARX and Back-Propagation models.
1. ERBIL
Erbil is located in the north of Iraq. It has a total land area of 438 320 km2 with a population of
28.8 million. Its latitude is 36.191113 and longitude is 44.009167. It is a mountainous region and
experience snowfall in winter season with temperature on average of 6 ͦC in urban and in the
mountainous ranges about -10 ͦC (MuttalebAlhashimi, 2014).
There are four seasons experienced in the north of Iraq and the winters are extremely cold. There
is heavy rainfall which continues for a long span of time and only a few months of the year have
very less or no rainfall. The weather of north Iraq has variations and the metrological
departments measure the parameters like humidity, air pressure, wind direction and wind speed
for predicting the rainfall daily, monthly or annual data.
28
Besides, the region of northern Iraq is rich in agricultural activities and production of agricultural
products associated with the need of rainfall prediction for effective and efficient supply and
production of agricultural products (Gibson, 2012). The irrigation system in north Iraq also needs
to focus on the reservation of water as it is a scares resource and being the agricultural economy;
it is also significant for this region to utilize water more effectively and efficiently (Frenken,
2009). This also makes it significant for the study to analyze the weather forecasting for this
region.
Figure 5.1: Map of Iraq; (a) Erbil (b) focus area of this study (north Iraq) (Zakaria, Al-Ansari,
Mustafa, & Knutsson, 2013).
2. NICOSIA
Nicosia is the capital city of North Cyprus. It has a total population 61,378 according to the
census of 2011. It has a total area of 3,355-kilometer square. The weather conditions of Nicosia
are extreme for summers and winters. The winters normally experience cold with mostly rain.
The latitude of Nicosia is 34⁰ and 36⁰ North and longitude of 32⁰ and 35⁰ East (SPO, 2014). It
29
experiences 60% of the annual rainfall from December to February. The weather in Nicosia is
usually very dry and humidity is experienced to be 56% (bbc).
Nicosia is an Urban/metropolitan city and occupies huge area; as it is the biggest cities in North
Cyprus. The city has a metrological department that predicts the rainfall to sustain the everyday
life and avoid the natural disasters. The rainfall parameters for the city of Nicosia are studied in
this particular research to evaluate the output prediction as that of Erbil in Iraq.
3. FAMAGUSTA
Famagusta is the seashore city of Northern Cyprus. It is located on the east coast of Northern
Cyprus. The city f Famagusta is not very huge like that of Nicosia but it has many beaches and
has a total population 46,900 according to the survey of 2015. The area of Famagusta consists of
35⁰7’13.94” of North and 33⁰56’20.18” of the East. The humidity in Famagusta is observed to
be (Akingbaso, 2014). The rainfall parameter data is collected for Famagusta as an input of this
study to evaluate the output for precision and accuracy.
Figure 5.2: Map of Northern Cyprus; showing Nicosia and Famagusta (Akingbaso, 2014).
30
5.1.1 Data Pre-Processing for Erbil
The monthly rainfall data from 2012-2017 for Erbil, is collected from the metrological
department to outline the trends and variations that are being observed within this time period
and to highlight the rain cycles for each that will effective for prediction.
Rainfull
200
150
100
50
0
Jan
Jan
Jan
Mar
Sep
Sep
Sep
Jan
Sep
May
Mar
Nov
May
Mar
Nov
May
Mar
Sep
Jan
Mar
Nov
May
Nov
May
Nov
Jul
Jul
Jul
Jul
Jul
Figure 5.3: Trends in the distribution of rainfall for Erbil
The figure 5.3 above is a graph showing the rainfall distribution of each month for Erbil. The
graph outlines that the cycle of the rainfall is varying for each year but the rain circle is constant
as observed. Each rain circle starts from November and ends in May that shows the winter
season in Erbil is very cold and rainy. Also, by the graph; it is prominent that distribution of
rainfall in winter season is high. Similarly, the high distribution of rain in winters indicates that
the summer season is comparatively dry with almost 0 mm rain as shown by the data. Therefore,
the weather of Erbil in summers is hot and dry. Furthermore, the data outline that for each year,
Erbil experienced showers for 7 months (Nov-May) continuously as significant amount of
rainfall is observed that is almost 180-185 mm, indicating that 5 months of the years are
extremely dry with no precipitation at all.
31
Tempreature
40
35
30
25
20
15
10
5
0
Jan
Jan
Sep
Mar
May
Nov
Mar
May
Sep
Jan
Jan
Nov
Mar
May
Sep
Jan
Nov
Mar
May
Sep
Nov
Mar
May
Sep
Nov
Jul
Jul
Jul
Jul
Jul
Figure 5.4: Monthly average temperature for Erbil
The above figure 5.4 highlights the average temperature data for Erbil. The graph outlines that
the average temperature is not changing for these years and similar patterns are observed with
the lowest temperature in January and highest temperature in August. Whereas, in January, the
city experiences the high precipitation and it’s cold and rainy winters with as low temperature as
5⁰ C but the hot and dry summer with average temperature of 35⁰C to 37⁰C. The trends in
average temperature are constant and the same pattern is observed from 2012-2017.
Humidity
100
80
60
40
20
0
Jan
Jan
May
Jan
Mar
May
Sep
Nov
Mar
Sep
Nov
Mar
May
Sep
Jan
Jan
May
Nov
Mar
May
Sep
Nov
Mar
Sep
Nov
Jul
Jul
Jul
Jul
Jul
Figure 5.5: Trends in humidity for Erbil
The above figure 5.5 outlines the trends in humidity for Erbil. The data in the humidity graph
outlines that highest rate of humidity is observed from the month of January up to May; the
32
rainfall period with low temperature. In comparison, the month of July outlines lowest humidity
rate because of no rainfall and extreme temperatures. The weather is extremely dry and hot
during this time period of each year as observed from the data. Also, the trends in humidity from
2012-2017 are not varying much.
windspeed
3.5
3
2.5
2
1.5
1
0.5
0
Sep
Jan
Jan
Mar
May
Nov
Mar
Sep
Jan
May
Nov
Mar
Sep
Jan
May
Nov
Mar
Sep
Sep
May
Nov
Jan
Mar
May
Nov
Jul
Jul
Jul
Jul
Jul
Figure 5.6: Average wind speed for Erbil
The exceeding figure 5.6 shows the average wind speed for Erbil from 2012-2017 which is not
varying except for the years 2014 and 2015 when the average wind speed is nearly zero. 2014
and 2015 were extremely hot in summers and almost dry throughout the year with the
precipitation for 2014 in February was 8.2 mm only and for February 2015, went to just 44.3
mm. 2014 and 2015 were the drought years that impact the production of agricultural products
and the agricultural industry directly. Because of nearly zero wind speed; there was no formation
of clouds that affected the rainfall vice versa.
Figure 5.7: Correlation between humidity and rainfall for Erbil
33
The figure 5.7 above shows the correlation between humidity and rainfall. It is expected that the
humidity increases with the increase in the amount of rainfall. For, Erbil the correlation is 0.76
which is quite good. Normally, with high relative humidity; the saturation of vapours into clouds
at a certain temperature helps in precipitation. The drier air will less accommodate the formation
of clouds and thus lead to low chances for the rain showers.
Figure 5.8: Correlation between temperature and rainfall for Erbil.
The figure 5.8 above illustrates that correlation between the temperature and rainfall using the
Plot Chart. The correlation between temperature and rainfall is 78.2% as outlined by the data.
This correlation is good for the prediction rainfall. As, increased temperature will affect the
rainfall consequently low temperature will compensate the precipitation and increase the amount
of rainfall.
Figure 5.9: Correlation between wind direction and rainfall for Erbil
The above figure 5.9 demonstrates the correlation between the wind direction and rainfall. The
correlation between wind direction and rainfall is 54% which is a very relationship.
34
The correlation data shows that there is an increased precipitation with the particular direction of
the wind.
Figure 5.10: Correlation between wind speed and rainfall for Erbil.
The above figure 5.10 highlights the correlation between the wind speed and rainfall. The data
reveals that there is only 10% correspondence between the wind speed and rainfall for the Erbil
city. As, the wind speed has no prominent role in predicting the rainfall for Erbil; this study will
least consider this parameter in rainfall prediction for Erbil. Since, the correlation obtained by
Plot Chart represents the least significant of wind speed with rainfall; considering it a noticeable
parameter is inappropriate.
35
Table 5.1: Correlation between input and outputs using SPSS for Erbil
The above table 5.1 illustrates the correlation between the inputs and outputs using the SPSS
software to find the inputs for training the Neural Network. Since, the data for air pressure is not
available that is why it is assigned the value of zero. The test results from SPSS demonstrate that
the main training inputs for Erbil to predict rainfall are: humidity, temperature and wind
direction.
5.1.2 Data Pre-Processing for Nicosia
The monthly rainfall data from 2012-2017 is collected for Nicosia as well. The parameters of the
rainfall that are to evaluate in this study are observed with their varying trends and are analyzed
for the forecasting.
36
Rain
120.0
100.0
80.0
60.0
40.0
20.0
0.0
Sep
Sep
Jan
Mar
May
Nov
Jan
Mar
Sep
May
Nov
Jan
Mar
May
Nov
Mar
Sep
Sep
Jan
May
Nov
Jan
Mar
May
Nov
Jul
Jul
Jul
Jul
Jul
Figure 5.11: Trends in distribution of rainfall for Nicosia
The above figure 5.11 shows the trends in the distribution of rainfall in the past six years for
Nicosia. The rainfall data highlights the changing rain cycles in Nicosia whereas, the overall
patterns are not varying much. Nicosia in the Northern Cyprus experiences rainfall usually in the
month of September to May. However, the data outline that the summers are hot and dry with
zero or very little rainfall. The winters are cold and rainy. Nicosia has a long-time period of rain
with only 4 months without precipitation. But the overall rain cycle in Nicosia is not constant;
sometimes it rains in September until November only.
temp
40
30
20
10
0
Jan
Jan
Jan
Mar
May
Sep
Nov
Mar
May
Sep
Nov
Mar
May
Sep
Jan
Nov
Mar
May
Sep
Jan
Nov
Mar
May
Sep
Nov
Jul
Jul
Jul
Jul
Jul
Figure 5.12: Average temperature for Nicosia
The above figure 5.12 illustrates the trends in average temperature for Nicosia. The average
temperature as shown in the graph is constant for the last years with lowest observed in February
and highest observed in July. There is no variation in the trends of average temperature that
37
would affect the rainfall. Also, the highest temperature goes to 30⁰C to 38⁰C as observed by the
monthly data.
humadity
90.0
80.0
70.0
60.0
50.0
40.0
30.0
20.0
10.0
0.0
Jan
Jan
Jan
Jan
Jan
Mar
Sep
May
Mar
Sep
Mar
Sep
Mar
Sep
Mar
Sep
Nov
May
Nov
May
Nov
May
Nov
May
Nov
Jul
Jul
Jul
Jul
Jul
Figure 5.13: Trends in humidity for Nicosia
The above figure 5.13 highlights the trends in the humidity for Nicosia. The graph shows that the
highest humidity rate is observed in the month of January whereas, the lowest is observed within
the month of July. As discussed above, the rainfall cycle starts in September with high humidity
observed within these months to compensate saturation of water vapours and formation of
clouds.
Wind Speed
40.0
30.0
20.0
10.0
0.0
Jan
Jan
Sep
Jan
Jan
May
Mar
May
Sep
Nov
Mar
May
Sep
Jan
Nov
Mar
May
Nov
Mar
May
Sep
Nov
Mar
Sep
Nov
Jul
Jul
Jul
Jul
Jul
Figure 5.14: Trends in wind speed for Nicosia
The above figure 5.14 demonstrates the changing trends in the wind speed for Nicosia. The wind
speed is a significant parameter to observe the rainfall and to make predictions for precipitation.
38
The wind speed is highest in September when the rainfall starts in Nicosia and is observed as
lowest in July when the rainfall ends normally. Although, there are variations in the overall wind
speed for these years but the average wind speed cycle is same.
Air pressure
1010.0
1005.0
1000.0
995.0
990.0
985.0
980.0
Jan
Jan
May
Jan
Jan
Mar
May
Sep
Nov
Mar
Sep
Nov
Jan
Mar
May
Sep
Nov
Mar
May
Sep
Nov
Mar
May
Sep
Nov
Jul
Jul
Jul
Jul
Jul
Figure 5.15: Trends in average air pressure for Nicosia
The above figure 5.15 illustrates the trends in the air pressure for Nicosia. The graph presents a
constant cycle of the air pressure observed with lowest in July when there is no rainfall and
highest in Dec/Jan that is the rainy season. Therefore, air pressure is also an important parameter
for predicting rainfall in Nicosia.
Wind direction
400.0
350.0
300.0
250.0
200.0
150.0
100.0
50.0
0.0
Sep
Sep
Jan
Mar
May
Sep
Nov
Jan
Mar
May
Nov
Sep
Jan
Mar
May
Nov
Jan
Mar
May
Nov
Sep
Jan
Mar
May
Nov
Jul
Jul
Jul
Jul
Jul
Figure 5.16: Trends in wind direction for Nicosia
The above figure 5.16 demonstrates the trends in the wind direction for Nicosia. The graph
shows that the wind direction is in favour of the rainfall and it observed especially in relation to
39
the rainfall season. This means that wind direction plays a significant role in the formation of
clouds and precipitation in Nicosia.
Figure 5.17: Correlation for rainfall with humidity and temperature
The above figures 5.17 illustrate the correlation between rain with humidity and rainfall and with
temperature respectively. The rainfall and humidity are observed with a close relation that is
47% and the correlation between rainfall and temperature are also high that is 52%. This means
that the humidity and temperature are significant parameters for predicting rainfall in Nicosia.
Figure 5.18: Correlation for rainfall with wind direction and rainfall with wind speed
The above figure 5.18 demonstrates the correlation between rainfall with wind direction and
rainfall and wind speed. The correlation is studied by using the Plot Chart and illustrates that
there is 35.3% correlation between the rainfall and wind speed which does not qualify this
40
parameter as a significant trainer for neural network. For the correlation between wind direction
and rainfall is observed to be 7.8% which is important in making rainfall predictions for Nicosia.
Table 5.2: Correlation between input and output using SPSS for Nicosia
The above table 5.2 illustrates the correlation between inputs and outputs using the SPSS
software. The correlation is analyzed for all these parameters and the result outlines that for
training the Neural Network to make rainfall predictions for Nicosia, temperature, humidity,
wind speed, and air pressure are most prominent for accuracy and precision.
5.1.3 Data Pre-Processing for Famagusta
The data for the rainfall parameters are analyzed for Famagusta in this study. The parameters of
rainfall are evaluated to be trained for neural-network to make predictions. The data for analysis
is gathered from the metrological department and analyzed for the seashore city. As the rainfall
parameters have a variance for the seashore city.
41
RAİN
400.0
350.0
300.0
250.0
200.0
150.0
100.0
50.0
0.0
Sep
Sep
Sep
Sep
Sep
Jan
Mar
May
Jan
Mar
May
Jan
Mar
May
Jan
Mar
May
Jan
Mar
May
Nov
Nov
Nov
Nov
Nov
Jul
Jul
Jul
Jul
Jul
Figure 5.19: Trends in distribution of rainfall in Famagusta
The above figure 5.19 illustrates the trends in the distribution of the rainfall in Famagusta. The
graph presents that the rainy season starts from September and ends in May as that of Nicosia.
The summer season is normally dry with no precipitation and the winters are usually cold and
wet. The data represents that the overall rain cycle is constant but there is variation in the
distribution of rainfall year to year.
TEMPRATURE
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0
Sep
Jan
Jan
Mar
May
Sep
Jan
Mar
May
Mar
May
Sep
Sep
Sep
Jan
Mar
May
Jan
Mar
May
Nov
Nov
Nov
Nov
Nov
Jul
Jul
Jul
Jul
Jul
Figure 5.20: Average temperature for Famagusta
The above figure 5.20 shows the trends in average temperature for Famagusta. The data indicates
that the lowest temperature is observed in the month of January and highest in the month of July.
The overall pattern of the temperature is constant from 2012 to 2017. The temperature affects the
42
rainfall in Famagusta, so this study will consider temperature into account for predicting rainfall
in Famagusta.
HUMIDITY
80.0
60.0
40.0
20.0
0.0
Jan
Jan
Jan
Jan
Jan
Mar
Sep
May
Mar
Sep
Mar
Sep
Mar
Sep
Mar
Sep
Nov
May
Nov
May
Nov
May
Nov
May
Nov
Jul
Jul
Jul
Jul
Jul
Figure 5.21: Trends in humidity for Famagusta
The above figure 5.21 shows the trends in the humidity for Famagusta. The humidity is observed
from the data as high throughout the year. Opposite to other two cities; the high humidity rate
throughout the year because Famagusta is a seashore city. Its climate is humid even when it’s hot
in the summer season. Therefore, for this study humidity is important for making rainfall
predictions in Famagusta.
WİND speed
40.0
30.0
20.0
10.0
0.0
Mar
Sep
Mar
Sep
Jan
Mar
Sep
Mar
Sep
Mar
Sep
Jan
May
Jan
May
May
Jan
May
Jan
May
Nov
Nov
Nov
Nov
Nov
Jul
Jul
Jul
Jul
Jul
Figure 5.22: Trends in average wind speed for Famagusta
The above figure 5.22 shows the trends in average wind speed for Famagusta. The average
pattern of the wind speed is constant with very few variations. The wind speed is highest for the
winter and rainy season that also favors the precipitation in Famagusta and low in the summer
season when there is zero or less rainfall.
43
Wind direction
400.0
350.0
300.0
250.0
200.0
150.0
100.0
50.0
0.0
Sep
Sep
Jan
Mar
Sep
Nov
Sep
Sep
May
Nov
Jan
Mar
May
Nov
Jan
Mar
May
Jan
Mar
May
Nov
Jan
Mar
May
Nov
Jul
Jul
Jul
Jul
Jul
Figure 5.23: Average wind direction for Famagusta
The above figure 5.23 shows the trends in the average wind direction for Famagusta. The wind
direction is in favor of the rainfall season that is from September to May. The wind direction is
therefore, an important parameter for predicting rainfall for Famagusta in this study.
air pressur
2500.0
2000.0
1500.0
1000.0
500.0
0.0
Jan
Jan
Mar
Sep
Jan
Sep
Jan
May
Mar
Sep
Jan
Nov
Mar
Mar
Sep
Mar
Sep
Nov
May
May
Nov
May
Nov
May
Nov
Jul
Jul
Jul
Jul
Jul
Figure 5.24: Trends in average air pressure for Famagusta
The above figure 5.24 represents the trends in the average air pressure for Famagusta. The air
pressure is constant throughout the years with single variation observed in February with 2000
Pascal. The normal air pressure as observed from data is 1000 Pascal. This means that the air
pressure is also a significant parameter for Famagusta’s prediction of rainfall.
44
Figure 5.25: Correlation for rainfall with humidity and temperature
The above figure 5.25 presents the correlation for the humidity with rainfall and rainfall with
temperature respectively. The correlation between humidity and rainfall is observed to be 24% in
Famagusta and 59% between temperature and rainfall. The correlations observed are closely
linked to the rainfall and therefore, this study will focus on humidity and temperature as inputs to
predict rainfall in Famagusta.
Figure 5.26: Correlation between rain and wind direction and rainfall with wind speed
The figure 5.26 above illustrates the correlation between rainfall and wind direction along with
its correlation with wind speed. The correlation between rainfall and wind direction is 44% and
that of wind direction is only 2%. The speed of the wind is closely linked to the rainfall
prediction and therefore is efficient input in this study.
45
Table 5.3: Correlation between inputs and outputs using SPSS for Famagusta
The above table 5.4 illustrates the correlation between the inputs and outputs using SPSS
statistics. The correlation is closely observed for the temperature, humidity, air pressure with
wind speed as significant for predicting the rainfall in Famagusta.
5.2 Flowchart for rainfall prediction
The below data flowchart outlines the selection of input and the procedure to get the output that
in this study is the rainfall prediction. The flowchart illustrates the first step as a selection of
inputs that are parameters, processing of these inputs and completing training, testing and
validation for accurate and precise output that is rain forecast.
46
5.3 Selection of the input and output data
In this study, the database is collected directly from the metrological departments to analyze the
inputs and outputs. The significant inputs are highlighted as average temperature, air pressure,
wind speed, wind direction and humidity and the output are rainfall for the years 2012-2017. The
rainfall season for Erbil is observed from November up to May; 7 months of rainfall in a year.
The rainfall season for Nicosia as highlighted by the data starts from September and ends in
May; that is around 9 months according to the data. For Famagusta, the rainfall season is similar
to that of Nicosia. Therefore, to predict rainfall for these three cities above mentioned parameters
will be considered as significant inputs to forecast precipitation.
47
5.4 Feature Extraction
This study is focusing on average temperature, air pressure, wind speed, wind direction and
humidity. The parameters are defined as:
1. TEMPERATURE
It is the degree to which something is hot or cold. It is affected by several factors like latitude,
altitude, and distance from the sea, wind with ocean current. It affects the rainfall in a way that
the amount of water vapour in the atmosphere determines the amount of rainfall as light or with a
heavy shower. Thus, at high temperatures there is heavy rainfall.
2. WIND DIRECTION
It is the origin of the wind from where it generates. The direction of the wind determines the
precipitation. Therefore, it is also important to analyze the direction of the wind as a parameter
for the rainfall prediction.
3. WIND SPEED
It is the velocity of the air moving from high pressure to low pressure. The wind speed is in close
correlation to the rainfall and it is necessary to consider it as an input for prediction. The increase
in the wind speed decreases the intensity of the rainfall. Therefore, to experience more
precipitation during the rainy season; the favourable wind speed is low rather than high.
4. HUMIDITY
It is the amount of water vapour present in the air for a given temperature that is significant for
saturation of vapours to form clouds. The increased moisture in the air will accommodate high
saturation. The humid weathers experience more precipitation than the dry weather.
5. AIR PRESSURE
The air pressure is the pressure within the atmosphere of the earth. The air pressure decreases
with the increase in the elevation from the earth surface. For example: the air pressure on
mountains is less as compared to plains.
48
5.5 Training, Testing and Validation
The creation of a database from inputs and outputs; the succeeding phase is the training the input
through back propagation algorithm, NARX and hybrid models. The training of the input data is
done by the MATLAB software by using the NNTOOL and ANFISTOOL. Training is preceding
step to testing of the input data with acceptable error is achieved. The NNTOOL automatically
retains 30% of the input data for testing and validation and 70% of input data for the training.
The ANFISTOOL requires manual retention of the input data for training, testing and validation.
The ANFISTOOL manually fed with 60% of the input for training and 40% for testing.
5.6 ANN
The artificial neural networks are multipurpose and feasible set of the system. These are
basically the computational methodologies that are used in diverse fields. The artificial neural
networks are based on elements called neurons and these neurons receive inputs for processing
the required results. The neural networks are multilayered nodes that are nonlinear and connected
by the weighted lines (Abraham, 2005). When the input is presented to the neural networks
during the training of the data; the successful training then offers the neural networks that are
capable of predicting the output. The neural networks are capable of classifying an object,
recognizing a particular pattern of multifactorial data, approaching a function and even
configuration (Mitchell, 1997). This study will focus on the learning algorithms like the feed-
forward backpropagation and NARX model to predict the rainfall using the inputs described
above.
5.6.2 Applying backpropagation and NARX model for Erbil
This study is focusing on the backpropagation for Erbil to obtain accuracy for training, testing
and validation of input and output data.
49
Figure 5.27: Proposed back propagation network architecture for Erbil
The above figure 5.27 represents the back-propagation network architecture for Erbil. The
architecture proposed 3 inputs for processing; temperature, humidity and wind direction. The
other parameters for predicting rainfall including wind speed and air pressure is not considered
as they are least correlated. The total number of neurons for this propagation is 100 with two
hidden layers. The output of this proposed back propagation network is rainfall. For the training
of this input data; Levenberg Marquardt (trainlm) which is one of the most effective and fast
algorithm functions in the toolbox, also it modernizes the weight and bias values with respect to
the Levenberg Marquardt and is used by the proposed back propagation network architecture
because of its high efficiency.
After training the data with 300 epochs, the result of the training, testing and validation is
obtained to check the accuracy as demonstrated in figure 5.28 below.
50
Figure 5.28: Snapshot of regression using NNTOOL
The accuracy of the training is 0.92 out of 1 which is absolutely precise and accurate result of the
training. Similarly, the accuracy of testing is 0.94 that outlines the accuracy and precision of the
tested data. Lastly, the accuracy of validation is 0.99 out of 1; that indicates the perfect result for
validation as well.
The proposed NARX network architecture has an input layer with two hidden layers and an
output layer. The network has a feedback connection for the neuron processing. There are three
inputs for this network; temperature, wind direction and humidity. The most suitable number of
neurons observed for this network is taken as 100. The output obtained is the predicted
precipitation for Erbil. The tapped delay lines TDL that are; 0:1 and 1:2 at the input side as
51
shown in the below figure 5.29, are supposed of length two. These are useful to sort the predicted
values during the training phase; concurrently reducing the complexity of the network.
Figure 5.29: Proposed NARX network architecture for Erbil
After training the data with the function of Levenberg Marquardt (trainlm), following results are
achieved. The gradient is equal to 0.45 at epoch 150. The Mu is equal to 0.01 out of 1.00e+10
and the validation checks (that are safeguard for overtraining and overfit of the net. Validation) is
equal to 150 at epoch 150 out of 1000 of max _fail, as shown in figure 5.30 below.
Figure 5.30: Training state for NARX
52
5.6.3 Applying backpropagation and NARX model for Nicosia
Similarly, this study focused on back propagation model for training, testing and validation for
Nicosia.
Figure 5.31: Proposed back propagation network architecture for Nicosia
The above figure 5.31 illustrates the proposed back propagation architecture for Nicosia. This
architecture is training with four parameters of rainfall in Nicosia that were significant in
predicting the precipitation. These input parameters include; temperature, wind speed, humidity
and air pressure and the output is rainfall. The proposed back propagation network architecture is
using 100 neurons for the input processing with one in out and one output layer. It has two
hidden layers and the training of the input is done using Gradient Descent with Momentum &
Adaptive LR (traingdx) algorithm. The traingdx is the function that trains the network that
upgrades weight and bias according to the gradient descent momentum and adaptive learning.
The algorithm is using 400 epochs for training the inputs and the output is rainfall that appears in
the output layer.
53
After training the network with 400 epochs, the results are obtained for training, testing and
validation. These results are checked for the accuracy and efficiency of the training network as
shown in the figure 5.32 above. The accuracy of training is 0.79 out of 1, the accuracy check for
testing is 0.94 and the validation accuracy is obtained as 0.96 out of 1. The results outline that
the accuracy of the training is precise and efficient by these four input parameters for Nicosia.
Figure 5.33: Proposed NARX network architecture for Nicosia
54
The above figure 5.33 demonstrates the proposed NARX network architecture for Nicosia. The
network uses 100 neurons and a feedback connection for processing the inputs. The network
function uses four input parameters for the training that are: temperature, wind speed, humidity
and air pressure. The function has one input layer and one output layer with two hidden layers
and the output obtained is the rainfall. The network architecture has tapped delay lines TDL; 0:1
and 1:2 that will help in predicting the values of the output during the training and helps in
simplification.
After training the data with Gradient Descent with Momentum & Adaptive LR (traingdx)
function. The network obtained the following results; the gradient is equal to 22.57 epochs 100
as shown in the figure 5.34 above. The Mu is equal to 0.01 out of 1.00e+10 and the validation
checks are equal to 100 at epoch 100 out of 1000 of max _fail.
55
5.6.4 Applying backpropagation and NARX model for Famagusta
The input and output parameters for Famagusta are also trained with back propagation model in
this study. The inputs are trained to check the accuracy of the output that is rainfall to validate
the predictions made through this network.
Figure 5.35: Proposed back propagation network architecture for Famagusta
The above figure 5.35 illustrates the proposed back propagation network architecture for
Famagusta. This propagation network is trained with four input parameters including;
temperature, humidity, and air pressure. The number of effective neurons is observed to be 100
for this training. The network has one input layer, one output layer for obtaining the output that
in this case is rainfall for Famagusta and two hidden layers. The Levenberg Marquardt (trainlm)
function is used for the processing of the input and output in this network.
After training the inputs and outputs with trainlm function; the following results are obtained as
shown in the figure 5.36 below. The result for training outlines accuracy equal to 0.64. Similarly,
the accuracy of the testing is 0.96 and lastly, the result of validation is equal to 0.81. These
results are obtained out of 1 that represents the enhanced accuracy and efficiency of the input
parameters to predict the rainfall in Famagusta. Therefore, temperature, humidity and air
pressure could be taken as reliable and effective input parameters to predict the rainfall in
Famagusta.
56
The figure 5.37 below represents the proposed NARX network architecture for Famagusta. The
NARX network uses four input parameters and the output is the rainfall. The NARX network
utilizes 100 neurons as effective for the prediction using (trainlm) function. There are two hidden
layers and the TDL 0:1 and 1:2 are expected to be of length 2 for an enhanced accuracy and to
minimize the complexity of the network.
Figure 5.37: Proposed NARX network architecture

57
The above figure 5.38 demonstrate the training state for NARX after training the inputs and
output, following results are achieved. The gradient is equal to 4789.49, the Mu is equal to 100 at
100 epochs and the validation checks are recorded as 100 at the total epoch of 100 out of 1000 of
max _fail, as shown above.
5.7 ANFIS
The ANFIS network is aimed at predicting the rainfall by using the data with inputs and outputs
from the past years. The prediction of the rainfall is a nonlinear system as discussed in this study
therefore, ANFIS is appropriate for making this prediction through training the algorithm, testing
with different functions and then checking. Subsequently, the model is tracked with several FIS
algorithms and with certain error tolerance and epochs to make the perfect predictions. The
parameters are analyzed through changing epochs and error tolerance to check the effect on
RMSE and to forecast rain. The model is aimed at minimizing the errors between the predicted
and the actual output that is rainfall.
58
5.7.1 Applying ANFIS for Erbil
For this study, the input and output parameters are trained using ANFIS for each city separately.
The input parameters are fed into the ANFIS algorithm manually, allocating 60% for training,
40% for testing the results obtained.
The below figure 5.39 illustrates the proposed ANFIS architecture for the city of Erbil. The
ANFIS model depends on the number of parameters and criterion. The ANFIS model has three
input criteria for Erbil that are: temperature, humidity and wind direction. The ANFIS model is
operated by using the hybrid algorithm with 100 epochs. 0.001 error tolerance is observed to be
suitable for the training. The architecture uses gauss2mf as the input from the MF type and linear
MF for the output. After training, testing and checking the ANFIS model with 3 inputs and 1
output; the error is obtained as 0.025 which is absolutely accurate and authentic for making the
prediction.
Figure 5.39: Proposed ANFIS architecture for Erbil
59
ANFIS after training the data with inputs and outputs, automatically generate the rules as shown
in the figure 5.40 below. After developing the rules by inputs, it gives output as rainfall
prediction. For each input it has different set of rules established.
Figure 5.40: ANFIS rule viewer for Erbil
5.7.2 Applying ANFIS for Nicosia
The ANFIS model for Nicosia is trained with four input criteria. The parameters are temperature,
humidity, air pressure and wind speed. The number of epochs is limited to 100 with the error
tolerance of 0.01 as shown in figure 5.38 below. The function of gauss2mf as the input and linear
MF for the output is selected for this model. After the training, testing and checking of the
parameters; the error is obtained as 0.011, indicating the precision and accuracy for the ANFIS
training of inputs and output.
60
Figure 5.41: Proposed ANFIS architecture for Nicosia
The figure 5.42 illustrates the AFIS rule viewer for Nicosia. The ANFIS used four inputs and the
output is rainfall prediction. The ANFIS generated different rules for the inputs for Nicosia.
Figure 5.42: ANFIS Rule viewer for Nicosia
5.7.3 Applying ANFIS for Famagusta
The below figure 5.41 demonstrate the proposed ANFIS model for Famagusta. The ANFIS
model is trained with four input criteria. The parameters are temperature, air pressure, humidity
61
and wind speed. The epochs are 100 for this training with an error tolerance of 0.01. The
function of gauss2mf as input is used and the linear MF for output is fixed as well. The error of
0.015 is obtained as a result of training, testing and checking. Therefore, ANFIS is also perfect
and most suitable model for forecasting precipitation to get accurate and precise data.
Figure 5.43: Proposed ANFIS architecture for Famagusta

The above figure 5.44 illustrate the ANFIS rule viewer for Famagusta. It is using four inputs for
Famagusta and predicted rainfall. The rules for Famagusta inputs are different from those of
Erbil and Nicosia.
Figure 5.44: ANFIS Rule Viewer for Famagusta
62
Table 5.4 below summarized the training, testing and validation for the data of Erbil, Nicosia and
Famagusta using backpropagation.
Table 5.4: Training, testing and validation by BPNN for Erbil, Nicosia and Famagusta
Place neurons epoch Inputs output Function Training testing validation
Erbil 100 300 temperature, Rainfall Trainlm 0.92 0.94 0.96

humidity and
wind direction.
Nicosia 100 400 temperature, Rainfall traingdx 0.79 0.94 0.96

wind speed,
humidity and
air pressure.
Famagusta 100 200 temperature, Rainfall Trainlm 0.64 0.96 0.89

humidity and
air pressure
Table 5.5 below highlights the summary of data trained, tested, and validated using NARX
model.
Table 5.5: Training, testing and validation by NARX for Erbil, Nicosia and Famagusta
Validation
Place Neurons Epoch Inputs output Function gradient Mu
Checks
temperature,
Erbil 100 150 humidity and Rainfall Trainlm 0.45 0.01 150
wind direction.
temperature,
wind speed,
Nicosia 100 100 Rainfall traingdx 22.5 0.01 100
humidity and
air pressure.
temperature,
Famagusta 100 200 humidity and Rainfall Trainlm 0.29 0.001 200
air pressure
63
Table 5.6 below outlines the training, testing, and validation of data using Hybrid model for
Erbil, Nicosia and Famagusta.
Table 5.6: Training, testing and validation by Hybrid for Erbil, Nicosia and Famagusta
Place MF type Epoch Inputs output Function Error Obtained

tolerance error
Erbil Linear 100 temperature, Rainfall gauss2mf 0.01 0.025

MF humidity and wind
direction.
Nicosia Linear 100 temperature, wind Rainfall gauss2mf 0.01 0.011

MF speed, humidity
and air pressure.
Famagusta Linear 100 temperature, Rainfall gauss2mf 0.01 0.015

MF humidity and air
pressure
64
CHAPTER 6
DISCUSSION AND RESULT
The study focused on using the artificial neural networks for predicting the rainfall on monthly
basis. The study proved that using back propagation and NARX with hybrid models can be used
to obtain accurate and precise forecast readings for rainfall, as demonstrated in this study.
Although, the networks were trained before testing and validation; the foremost challenge of the
study was to achieve maximum accuracy and efficiency in making the prediction.
Thus, the study trained and tested the monthly data for last 6 years of three different locations
from two different regions. For this purpose, the study focused on neural networks collaborated
with neuro-fuzzy. The study tested the rainfall data from 2012-2017. The areas for the data
collection and data testing were: Erbil, Nicosia and Famagusta using MATLAB software with its
NNTOOL and ANFISTOOL
6.1 Comparing results
Table 6.1: Obtained RMSE after training and testing
Location No. of neurons RMSE RMSE
BPNN NARX HYBRID
Erbil 100 3.7554 15.6513 0.025
Nicosia 100 8.8172 13.4775 0.011
Famagusta 100 57.3014 85.8441 0.015
The above table 6.1 illustrates the root mean square error obtained for Erbil, Nicosia and
Famagusta. The table demonstrates that the most effective result is obtained for Nicosia using
hybrid algorithm of ANFIS model with RMSE equal to 0.011.
65
6.2 Actual and predicted data for Erbil
Predicted Data By BPNN

200
180
160
140
120
100
80
60
40
20
0
rainfall Predict BPNN
Figure 6.1: Comparing actual and predicted data for Erbil using BPNN
The above figure 6.1 illustrates the graph comparing actual and predicted data for Erbil using
BPNN. The graph demonstrates that there is some difference between the actual and the
predicted data using BPNN. Similarly, the graph below highlights the comparison between the
actual and predicted data using NARX that highlights there is a more difference as show in
figure 6.3 below.
Predicted Data By NARX

400
300
200
100
rainfall Predict NARX
Figure 6.2: Comparing actual and predicted data for Erbil using NARX
66
Predicted Data By ANFIS
400
200
0
-200
-400
-600
-800
-1000
rainfall Predict Anfis
Figure 6.3: Comparing actual and predicted data for Erbil using Hybrid
The above figure 6.3 illustrates the comparison between actual and predicted data as observed by
using hybrid algorithm. The graph demonstrates a perfect result for actual and predicted data as
that of Erbil. The data shows accuracy and precision in the output data obtained that is the
rainfall prediction. The RMSE for hybrid was only 0.025 as discussed above in table 6.1.
6.3 Actual and predicted data for Nicosia

120
100
80
60
40
20
0
Figure 6.4: Comparing actual and predicted data for Nicosia using BPNN
67
The above figure 6.4 illustrates the graph comparing actual and predicted data for Nicosia using
BPNN. The graph highlights that there is a not much similarity and accuracy between the actual
and predicted data. Thus, the BPNN is not much efficient for Nicosia as well.

120
100
80
60
40
20
Figure 6.5: Comparing actual and predicted data for Nicosia using NARX
The above figure 6.5 demonstrates the graph of predicting data in comparison to the actual data
using NARX. There is a huge difference between actual and predicted data and accuracy is far to
achieve.

120
100
80
60
40
20
0
-20
Figure 6.6: Comparing actual and predicted data for Nicosia using Hybrid
68
The above figure 6.6 illustrates the comparison between actual and predicted data using Hybrid
model. The graph outlines that there is a close relation between the actual and predicted data
with maximum accuracy.
6.4 Actual and predicted data for Famagusta

400
350
300
250
200
150
100
50
0
Figure 6.7: Comparing actual and predicted data for Famagusta using BPNN
The above figure 6.7 illustrates the graph making comparison between the actual and predicted
data using BPNN for Famagusta. The graph outlines that there is a much difference between the
actual and predicted data throughout the graph.

400
300
200
100
Figure 6.8: Comparing actual and predicted data for Famagusta using NARX
69
The figure 6.8 above represents the comparison between the actual and predicted data for
Famagusta using NARX. The graph illustrates that there is just some similarity between the
actual and predicted data.

400
300
200
100
-100
Figure 6.9: Comparing actual and predicted data for Famagusta using Hybrid
The above figure 6.9 illustrates the comparison between the actual and predicted data for
Famagusta using the Hybrid model. The graph shows that the actual and predicted data is much
similar and accurate.
70
CONCLUSION
Rainfall is one the most significant natural phenomenon that is not only important for the human
beings only but the living beings. Due to the changing climatic conditions, rainfall cycles are
also changing and the temperature of the earth is rising. The changing temperature is also
affecting the agriculture, industry and sometimes may cause flooding and land slide. Therefore, it
is essential for the human beings to keep a check upon this natural phenomenon in order to
survive. The water is a scarce natural resource without which human life is impossible and also
there is no substitute to this natural resource. Thus, predicting the rainfall for agriculture and
water reserves, also it also good for keeping human beings alert of natural disasters like flood
and landslide. However, to overcome these issues and meet the demands, a system to forecast
rainfall is essential using artificial intelligence of neural that is popular within the modern
technology.
The study aimed at building a predicting system using neural networks that could predict
monthly rainfall accurately and efficiently with minimum error. The study incorporated different
areas and used their rainfall data with different neural networks like ANFIS and ANN, through
training the networks with these inputs and outputs. The trained data is tested and then validated
by making a comparison between actual and predicted data. The system used feature extraction
to deduce the output prediction that could be more precise and accurate. The neural networks
with different algorithms and functions were trained with rainfall parameters and the previous
rainfall data to predict the results in this study. After training and testing; the results were
compared to check the efficiency of the system; the RMSE’s were recorded to make sure that the
system will operate not only to make the prediction but also the accurate data will be obtained.
The study utilized back propagation, NARX and Hybrid algorithms to forecast the rainfall.
Lastly, the rainfall predictions after training, testing are obtained that are quite accurate and
through comparison outlined that the actual and predicted data for these areas illustrated finest
results using the certainly different parameters of the rainfall that are different for different areas
with minimum error observed using ANFIS only. The AFIS model outlined efficiency for Erbil,
Nicosia and Famagusta. The NARX network performed not so well in the comparison for actual
and predicted data.
71
REFERENCES
Abhishek, K., Kumar, A., Ranjan, R., & Kumar, S. (2012). A Rainfall Prediction Model using
Artificial Neural Network. IEEE Control and System Graduate Research Colloquium, 1-
5.
Abraham, A. (2005). Artificial Neural Networks. John Wiley and Sons, Ltd.
Adler, R. F., Huffman, G. J., Bolvin, D. T., Curtis, S., & Nelkin, E. J. (2000). Tropical Rainfall
Distributions Determined Using TRMM Combined with Other Satellite and Rain Gauge
Information. American Metrological Society, 1-9.
Agnihotri, G., & Panda, J. (2014). Comparison of Rainfall from Ordinary and Automatic Rain
Gauges in Karnataka. Mausam, 65(4), 1-8.
Ahn, J. (2017). Analysis of a neural network model for building energy hybrid controls for in-
between season. Architecture of Complexity, (pp. 1-5).
Akingbaso, E. Y. (2014). Land Use - Cover Change Assessment Framework: Famagusta North
Cyprus. Approval of the Institute of Graduate Studies and Research-EMU.
Alpers, W., & Melsheimer, C. (2004). Rainfall. SAR Marine User Manual, US Dept. of
Commerce, NOAA.
Amoo, O. T., & Dzwairo, B. (2016). Trend analysis and artificial neural networks forecasting for
rainfall trends. Environmental Economics, 7(4), 1-10.
bbc. (n.d.). Weather. Retrieved December 11, 2017, from cypnet.co,uk:

http://www.cypnet.co.uk/ncyprus/main/weather/index.html
Beard, J. S. (1962). Rainfall interception by grass. South African Forestry Journal, 42(1), 1-7.
Biswas, S. K., Marbaniang, L., Purkayastha, B., Chakraborty, M., Singh, H. R., & Bordoloi, M.
(2016). Rainfall forecasting by relevant attributes using artificial neural networks - a
comparative study. International JOurnal of Big Data Intelligence, 3(2), 1-10.
72
Blyth, A. M., Bennett, L. J., & Collier, C. G. (2015). High-resolution observations of
precipitation from cumulonimbus clouds. Meteorological applications, 22, 1-12.
Carvalho, J. R., Assad, E. D., Oliveira, A. F., & Pinto, H. S. (2014). Annual maximum daily
rainfall trends in the Midwest, southeast and southern Brazil in the last 71 years. Weather
and Climate Extremes, 5(6), 1-7.
Cheng, M., & Qi, Y. (2001). Frontal Rainfall-Rate Distribution and Some Conclusions on the
Threshold Method. Journal of applied meteorology, 41, 1-8.
Collier, C. G. (2003). On the formation of stratiform and convective cloud. Weather, 58, 1-4.
Darji, M. P., Dabhi, V. K., & Prajapati, H. B. (2015). Rainfall Forecasting Using Neural
Network: A Survey. International Conferences on Advances in Computer Engineering
and Applications (pp. 1-5). Ghaziabad, India: IEEE.
Dayan, U., Ziv, B., Margalit, A., Morin, E., & Sharon, D. (2001). A severe autumn storm over
the middle-east: synoptic and mesoscale convection analysis. Theoretical and Applied
Climatology, 69(1-2), 1-8.
Demuth, H. B., Beale, M. H., Jess, O. D., & Hagan, M. T. (2014). Neural Network Design.
Martin Hagan.
Devi, S. R., Arulmozhivarman, P., Venkatesh, C., & Agarwal, P. (2016). Performance
Comparison of Artificial Neural Network Models for Daily Rainfall Prediction.
International Journal of Automation and Computing, 13(5), 1-9.
Fang, L., & Zhongda, L. (2015). Improving Multi-model Ensemble Probabilistic Prediction of
Yangtze River Valley Summer Rainfall. Advances in atmospheric sciences, 32, 1-7.
Faulina, R., & Suhartono. (2013). Hybrid ARIMA-ANFIS for Rainfall Prediction in Indonesia.
International Journal of Science and Research, 2(2), 1-3.
Frenken, K. (2009). Irrigation in the Middle East region in figures – AQUASTAT Survey. Rome:
Food and Agriculture Organization of the United Nations (FAO).
73
Gibson, G. R. (2012). War and Agriculture: Three Decades of Agricultural Land Use and Land
Cover Change in Iraq. Blacksburg, VA: Diss.
Gray, W. R., & Seed, A. W. (2006). The characterisation of orographic rainfall. Metrological
Applications, 7(2), 1-12.
Gu, G., Adler, R. F., Huffman, G. J., & Curtis, S. (2007). Tropical rainfall variability on
interannual-to-interdecadal/longer-time scales derived from the GPCP monthly product.
Journal of Climate, 1-12.
Hansen, A. (1961). A Self‐Recording Electronic Osmometer for Quick, Direct Measurement of

Colloid Osmotic Pressure in Small Samples. Acta Physiologica, 53(3-4), 1-5.
Haurwitz, B., & Austin, J. M. (1944). Climatology. New York; London: Mcgraw-Hill Book
Company.
Hoai, N. D., Udo, K., & Mano, A. (2011). Downscaling Global Weather Forecast Outputs Using
ANN for Flood Prediction. Journal of Applied Mathematics, 1-12.
Hung, N. Q., Babel, M. S., Weesakul, S., & Tripathi, N. K. (2009). An artificial neural network
model for rainfall forecasting in Bangkok, Thailand. Hydrology and Earth System
Sciences, 13(8), 1-9.
Johnson, D. H. (1962). Rain in East Africa. Quarterly Journal of the Royal Metrological Society,
88(375), 1-10.
Jr., R. A. (2012). Orographic effects on precipitating clouds. Reviews of Geography, 50(1), 1-17.
Karimi, S., Shiri, J., Kisi, O., & Shiri, A. A. (2016). Short-term and long-term streamflow
prediction by using 'wavelet–gene expression' programming approach. ISH Journal of
Hydraulic Engineering, 22(2), 1-15.
Khamis, A., Nabilah, S., & Abdullah, S. B. (2014). Forecasting Wheat Price Using
Backpropagation And NARX Neural Network. The International Journal Of Engineering
And Science, 3(11), 1-7.
74
Khan, M. Z., Sharma, A., Mehrotra, R., Schepen, A., & Wang, Q. J. (2015). Does improved
SSTA prediction ensure better seasonal rainfall forecasts? Water Resources Research, 51,
1-12.
Kumar, A., Yang, F., Goddard, L., & Schubert, S. (2004). Differing Trends in the Tropical
Surface Temperatures and Precipitation over Land and Oceans. Journal of Climate,
17(3), 1-12.
Kumarasiri, A., & Sonnadara, D. (2006). Rainfall Forecasting: An Artificial Neural Network
Approach. Proceedings of the Technical Sessions (pp. 1-9). Institute of Physics – Sri
Lanka.
Li, P.-H., Kwon, H.-H., Sun, L., Lall, U., & Kao, J.-J. (2009). A Modified Support Vector
Machine Based Prediction Model on Streamflow at the shihmen Reservoir, Taiwan.
International Journal of Climatology, 30, 1-8.
Lima, P. M., & Guedes, E. B. (2015). Rainfall Prediction for Manaus, Amazons with Artificial
Neural Networks. Computational Intelligence (pp. 1-5). Curitiba, Brazil: IEEE.
McGREGOR, G. R., & PHILLIPS, I. D. (2003). Specification and prediction of monthly and
seasonal rainfall over the south-west peninsula of England. Q.J.R. Metrological Society,
130, 1-12.
Mitchell, T. M. (1997). Artificial Neural Networks. Machine Learning, 45, 1-12.
Mittal, P., Chowdhury, S., Roy, S., Bhatia, N., & Srivastav, R. (2012). Dual Artificial Neural
Network for Rainfall-Runoff Forecasting. Journal of Water Resource and Protection,
4(12), 1-4.
MuttalebAlhashimi, S. A. (2014). Prediction of Monthly Rainfall In Kirkuk Using Artificial

Neural Network and Time Series Model. Journal of Engineering and Development,
18(1), 1-12.
Narvekar, M., & Fargose, P. (2015). Daily Weather Forecasting using Artificial Neural Network
Neural Network. International Journal of Computer Applications, 121(22), 1-4.
75
NASA. (2005, December 19). Retrieved December 9, 2017, from eoimages.gsfc.nasa.gov:
https://eoimages.gsfc.nasa.gov/images/imagerecords/6000/6151/amsterdam_tmo_200535
3_lrg.jpg
Nayak, D. R., Mahapatra, A., & Mishra, P. (2013). A Survey on Rainfall Prediction using
Artificial Neural Network. International Journal of Computer Applications, 72(16), 1-14.
Nystuen, J. A. (1990). A note on the attenuation of surface gravity waves by rainfall. Journal of
Geophysical Research: Oceans, 95(C10), 1-2.
Omotosho, J. B., Balogun, A., & Ogunjobi, K. (2000). Predicting monthly and seasonal rainfall,
onset and cessation of the rainy season in west africa using only surface data.
International journal of climatology, 20, 1-16.
Qian, W., Kang, H.-S., & Lee, D.-K. (2002). Distribution of seasonal rainfall in the East Asian
monsoon region. Theoretical and Applied Climatology, 73(3-4), 1-9.
Robertson, D. E., Shrestha, D. L., & Wang, Q. J. (2013). Post-processing rainfall forecasts from
numerical weather prediction models for short-term streamflow forecasting. Hydroloy
and Earth System Science, 17, 1-13.
Rodgers, E. B., & Adler, R. F. (1980). Tropical Cyclone Rainfall Characteristics as Determined
from a Satellite Passive Microwave Radiometer. Monthly Weather REview, 109, 1-6.
Roe, G. H. (2005). Orographic precipitation. Annual Review of Earth Planetary Science, 33, 1-
18.
Schmidt, J., Turek, G., Clark, M. P., Uddstrom, M., & Dymond, J. R. (2008). Probabilistic
forecasting of shallow, rainfall-triggered landslides using real-time numerical weather
predictions. Natural Hazards and Earth System Sciences, 8(2), 1-7.
Selase, A. E., Agyimpomaa, D. E., Selasi, D. D., & Hakii, D. M. (2015). Precipitation and
Rainfall Types with Their Characteristic Features. Journal of Natural Sciences Research,
5(20), 1-3.
76
Selase, A. E., Agyimpomaa, D. E., Selasi, D. D., & Hakii, D. M. (2015). Precipitation and
Rainfall Types with Their Characteristic Features. Journal of Natural Sciences Research
www.iiste.org, 5(20), 1-3.
Shaikh, L., & Sawlani, K. (2017). A Rainfall Prediction Model Using Articial Neural Network.
International Journal of Technical Research and Applications, 5(2), 1-3.
Sharma, A., & Nijhawan, G. (2015). Rainfall Prediction Using Neural Network. International
Journal of Computer Science Trends and Technology, 3(3), 1-4.
Singhrattna, N., Rajagopalan, B., Clark, M., & Kumar, K. K. (2005). Seasonal forecasting of
thailand summer monsoon rainfall. International journal of climatology, 25, 1-14.
Sivakumar, M. V. (2006). Climate prediction and agriculture: current status and future
challenges. Climate research, 33(3-17), 1-12.
Skorpil, V., & Stastny, J. (2006). Neural networks and back propagation algorithm. Electronics,
1-4.
Smith, R. B., & Evans, J. P. (2007). Orographic Precipitation and Water Vapor Fractionation
over the Southern Andes. Journal of hydrometeorology, 8, 1-8.
Smola, A., & Vishwanathan, S. (2008). Introduction to Machine Learning. Cambridge: The
Press Syndicate of the University of Cambridge.
SPO. (2014). Economic and social indicators. TRNC.
Thatcher, L., Takayabu, Y. N., Yokoyama, C., & Pu, Z. (2012). Characteristics of tropical
cyclone precipitation features over the western Pacific warm pool. Journal of geophysical
research, 117, 1-8.
Trenberth, K. E. (2011). Changes in precipitation with climate change. Climate research, 47, 1-
14.
Wahyuni, I., Mahmudy, W. F., & Iriany, A. (2017). Rainfall Prediction Using Hybrid Adaptive
Neuro Fuzzy Inference System (ANFIS) and Genetic Algorithm. Journal of
Telecommunication, Electronic and Computer Engineering, 9(2-8), 1-6.
77
White, C. J., Franks, S. W., & McEvy, D. (2015). Using subseasonal-to-seasonal (S2S) extreme
rainfall forecasts for extended-range flood prediction in Australia. Proceedings of the
International Association of Hydrological Sciences (pp. 1-5). Copernicus Publications on
behalf of the International Association of Hydrological Sciences.
Wichitarapongsakun, P., Sarin, C., Klomjek, P., & Chuenchooklin, S. (2016). Rainfall prediction
and meteorological drought analysis in the Sakae Krang River basin of Thailand.
Agriculture and Natural Resources, 50(6), 1-7.
Y.H.Zweiri, J.F.Whidborne, & L.D.Seneviratne. (2002). A three term back propagation

algorithm. Neurocomputing, 50, 1-13.
Yegnanarayana, B. (2009). Artificial neural networks. PHI Learning Pvt. Ltd.
Zakaria, S., Al-Ansari, N., Mustafa, Y. T., & Knutsson, S. (2013). Macro Rain Water Harvesting
Network to Estimate Annual Runoff at Koysinjaq (Koya) District, Kurdistan Region of
Iraq. Engineering, 5(12), 1-8.
Zealand, C. M., Burn, D. H., & Simonovic, S. P. (1999). Short term streamflow forecasting using
artificial neural networks. Journal of Hydrology, 214, 1-15.
78
APPENDICES
79
APPENDIX A
DATABASE FOR ERBIL
Month Year Temperature Humidity Wind direction Windspeed Rainfall Air pressure
Jan 2012 8.2 74.4 120 1.5 56.1 0.0
Feb 2012 9.55 63.6 160 2.3 31.5 0.0
Mar 2012 11.2 61.7 150 1.9 72.1 0.0
Apr 2012 21.95 49 190 2.9 14.5 0.0
May 2012 27.1 41.5 270 2.4 26.8 0.0
Jun 2012 33.15 25.8 270 2.8 0 0.0
Jul 2012 35.9 22.1 270 2.5 0 0.0
Aug 2012 35.15 28.5 220 1.8 0 0.0
Sep 2012 30.8 29 90 1.9 0.2 0.0
Oct 2012 24.45 48.6 120 2 23.4 0.0
Nov 2012 17.15 64.6 60 1.7 45.9 0.0
Dec 2012 6 77.2 150 1.7 95.9 0.0
Jan 2013 7.45 73.9 140 2.2 174.4 0.0
Feb 2013 12.05 75.6 160 1.6 55.8 0.0
Mar 2013 14.95 61.6 140 1.9 17.7 0.0
Apr 2013 20.35 53.9 180 1.8 37.4 0.0
May 2013 25.3 47.7 290 1.9 40.6 0.0
Jun 2013 31.4 31.2 270 1.4 0 0.0
Jul 2013 34.3 29.1 270 1.2 0 0.0
Aug 2013 34.05 28.7 270 1.1 0 0.0
Sep 2013 28.8 37.8 270 0.8 0 0.0
Oct 2013 23.2 39.4 90 1.5 0.2 0.0
Nov 2013 17.4 67.9 90 1.1 19.1 0.0
Dec 2013 9.6 66 90 1.1 86.6 0.0
Jan 2014 10.6 72.1 90 1 74.9 0.0
Feb 2014 11.5 56 90 1.4 8.2 0.0
Mar 2014 15.8 66.2 120 2 93.4 0.0
Apr 2014 20.6 48.6 270 0.8 14.7 0.0
May 2014 27.3 39.8 270 0.2 2.5 0.0
Jun 2014 31.8 30.3 270 0.1 0 0.0
Jul 2014 35 29.1 270 0.1 0 0.0
Aug 2014 35.3 28.3 270 0.1 1 0.0
Sep 2014 29.6 39.8 270 0.1 0.0 0.0
Oct 2014 22.8 58.6 90 0.1 68.7 0.0
Nov 2014 14.6 65.6 100 0.2 69.2 0.0
80
Dec 2014 11.7 74.7 90 0.2 56.0 0.0
Jan 2015 9.05 72.5 90 0.2 37.4 0.0
Feb 2015 11 68.4 90 0.1 44.3 0.0
Mar 2015 14.2 64.3 90 0.2 56.8 0.0
Apr 2015 19.2 52.2 270 0.2 17.3 0.0
May 2015 27.4 34.5 270 0.2 5.4 0.0
Jun 2015 31.5 33.2 270 1.7 0 0.0
Jul 2015 36.15 25.4 270 1.3 0 0.0
Aug 2015 35.45 30.8 270 1.3 0 0.0
Sep 2015 32.25 29.9 90 1.9 4.5 0.0
Oct 2015 24.6 54.6 270 2.3 36.4 0.0
Nov 2015 15.05 69.8 90 1.3 104.4 0.0
Dec 2015 9.9 67.7 90 1.5 115.8 0.0
Jan 2016 7.9 76.4 270 1.4 45.6 0.0
Feb 2016 12.15 65.3 90 1.5 33.0 0.0
Mar 2016 17.6 55.2 90 1.9 103.5 0.0
Apr 2016 19.65 53.9 90 1.7 46.4 0.0
May 2016 26.1 40.5 270 1.7 6.4 0.0
Jun 2016 32.55 32 90 2.2 2.5 0.0
Jul 2016 35.45 27.3 270 1.2 0 0.0
Aug 2016 36.15 30.5 270 1.1 0 0.0
Sep 2016 29.5 39.4 270 1.2 0 0.0
Oct 2016 24.5 41 270 1.4 T.R 0.0
Nov 2016 15.55 47.5 90 1.1 22.3 0.0
Dec 2016 8.65 76.4 150 1.5 116.1 0.0
APPENDIX B
DATABASE FOR NICOSIA
Month Year Temp Humidity Wind speed Wind direction Air pressure Rainfall
Jan 2012 9.0 74.8 18.5 360.0 999.5 86.0
Feb 2012 8.4 68.0 19.9 270.0 1000.2 53.6
Mar 2012 11.3 61.1 20.7 90.0 1001.7 12.6
Apr 2012 16.9 59.6 19.9 315.0 997.1 10.8
May 2012 20.8 60.0 18.2 315.0 996.1 70.4
Jun 2012 26.9 52.0 19.8 90.0 992.9 0.60
Jul 2012 30.4 46.2 19.6 360.0 989.4 0.0
Aug 2012 30.0 44.7 17.5 360.0 991.2 5.4
81
Sep 2012 26.2 54.9 12.8 270.0 995.1 0.0
Oct 2012 21.6 63.8 30.8 45.0 998.4 73.2
Nov 2012 17.0 68.2 14.2 360.0 1000.4 35.0
Dec 2012 12.0 77.7 20.0 370.0 999.9 60.3
Jan 2013 9.6 76.0 16.6 360.0 1000.5 63.2
Feb 2013 11.3 75.5 17.5 315.0 1000.6 11.0
Mar 2013 13.4 61.2 24.6 360.0 997.9 8.6
Apr 2013 17.7 57.0 18.2 360.0 997.2 28.8
May 2013 23.1 52.9 16.6 90.0 995.4 90.6
Jun 2013 26.2 47.6 21.0 360.0 993.2 0.0
Jul 2013 28.6 47.8 15.3 360.0 990.5 0.2
Aug 2013 29.5 47.0 13.9 360.0 991.1 0.0
Sep 2013 25.4 52.1 17.2 360.0 995.1 0.0
Oct 2013 20.0 41.0 17.8 45.0 1000.4 2.0
Nov 2013 17.3 63.9 14.8 90.0 1000.3 20.8
Dec 2013 10.2 63.1 24.4 360.0 1000.4 37.0
Jan 2014 11.6 76.0 14.6 180.0 1003.1 11.4
Feb 2014 11.6 75.5 15.7 360.0 1002.5 15.0
Mar 2014 14.5 61.2 24.6 360.0 998.1 30.8
Apr 2014 17.7 57.0 16.5 315.0 998.1 38.6
May 2014 20.9 52.9 15.5 315.0 996.7 41.2
Jun 2014 26.1 47.6 17.6 270.0 994.3 0.0
Jul 2014 28.5 47.8 16.2 270.0 991.7 0.0
Aug 2014 29.2 47.0 13.3 247.5 991.3 0.4
Sep 2014 25.6 52.1 15.2 270.0 995.1 11.4
Oct 2014 20.5 41.0 22.0 135.0 999.4 89.0
Nov 2014 14.7 63.9 13.6 90.0 1001.7 26.2
Dec 2014 12.9 63.1 18.1 360.0 1003.4 61.8
Jan 2015 9.2 75.6 17.3 270.0 1002.7 46.8
Feb 2015 9.7 75.8 17.3 45.0 997.8 44.6
Mar 2015 13.1 68.9 20.7 360.0 1001.0 48.8
Apr 2015 15.6 58.5 20.4 315.0 998.8 19.6
May 2015 21.8 56.3 16.2 270.0 996.2 37.0
Jun 2015 24.7 54.5 15.4 315.0 994.3 0.0
Jul 2015 28.7 49.0 13.9 360.0 992.1 0.2
Aug 2015 29.7 50.0 14.7 315.0 993.3 0.2
Sep 2015 27.0 57.1 14.3 45.0 996.1 22.4
Oct 2015 22.3 62.6 14.9 360.0 998.6 4.4
Nov 2015 16.4 55.6 17.0 360.0 1002.0 2.6
Dec 2015 11.3 62.4 19.2 360.0 1006.9 13.2
Jan 2016 9.3 69.0 19.8 360.0 1003.1 26.2
82
Feb 2016 13.2 64.6 19.8 315.0 1003.6 13.6
Mar 2016 14.4 59.4 22.6 270.0 998.2 47.4
Apr 2016 20.1 50.8 18.1 360.0 997.2 17.4
May 2016 21.7 51.8 18.2 315.0 996.1 30.0
Jun 2016 28.0 44.1 16.9 270.0 993.4 0.0
Jul 2016 30.2 46.4 20.3 135.0 990.6 0.0
Aug 2016 29.4 53.2 14.8 90.0 992.7 0.0
Sep 2016 25.8 52.1 16.4 270.0 996.4 5.2
Oct 2016 22.3 54.3 16.6 292.5 999.3 0.2
Nov 2016 15.4 55.2 19.3 360.0 1003.6 16.8
Dec 2016 9.9 72.8 20.0 315.0 1003.2 110.6
APPENDIX C
DATABASE FOR FAMAGUSTA
Month Year Temperature Humidity Wind speed Wind Air pressure Rainfall
direction
Jan 2012 12.2 70.5 23.3 225.0 1015.1 235.0
Feb 2012 12.2 64.0 22.8 270.0 1014.9 63.0
Mar 2012 13.0 65.2 23.2 90.0 1016.9 37.4
Apr 2012 17.2 71.7 19.8 315.0 1012.2 4.8
May 2012 20.8 72.4 19.9 315.0 1011.0 39.6
Jun 2012 25.4 70.5 14.4 225.0 1007.6 0.2
Jul 2012 29.0 64.1 17.8 360.0 1003.9 0.0
Aug 2012 29.1 60.8 18.3 225.0 1005.6 2.8
Sep 2012 26.2 65.8 15.3 225.0 1009.7 0.0
Oct 2012 23.1 68.0 21.1 180.0 1013.0 51.6
Nov 2012 19.6 66.7 17.8 225.0 1015.2 227.6
Dec 2012 15.1 70.3 25.0 315.0 1015.1 108.4
Jan 2013 13.0 69.6 21.8 360.0 1015.4 109.8
Feb 2013 14.2 71.8 25.4 315.0 1015.8 45.2
Mar 2013 15.3 64.0 29.2 315.0 1013.1 26.6
Apr 2013 18.0 65.0 21.6 315.0 1012.2 17.0
May 2013 22.5 66.0 17.2 225.0 1010.2 22.6
Jun 2013 25.4 62.4 18.5 225.0 1007.8 0.0
Jul 2013 27.7 61.4 15.8 225.0 1005.0 0.0
Aug 2013 28.6 61.6 16.0 225.0 1005.6 0.0
Sep 2013 25.9 65.0 18.3 45.0 1009.8 0.0
Oct 2013 21.2 52.3 20.3 315.0 1015.2 4.0
83
Nov 2013 19.7 65.9 18.0 45.0 1015.2 41.0
Dec 2013 13.6 62.9 27.5 90.0 1019.5 89.0
Jan 2014 14.6 68.7 18.1 45.0 1018.5 58.2
Feb 2014 13.7 67.1 22.9 45.0 1017.8 27.4
Mar 2014 15.8 64.3 24.3 360.0 1013.2 27.6
Apr 2014 18.2 67.5 20.7 315.0 1013.2 12.6
May 2014 21.0 68.8 16.1 225.0 1011.8 54.8
Jun 2014 25.1 60.4 18.7 360.0 1009.0 4.4
Jul 2014 27.5 68.7 28.3 225.0 1006.4 0.0
Aug 2014 28.2 71.2 17.8 225.0 1006.8 0.0
Sep 2014 26.4 61.3 21.4 315.0 1007.8 2.2
Oct 2014 22.1 64.8 25.6 90.0 1013.3 54.6
Nov 2014 17.8 62.8 18.9 360.0 1016.9 34.4
Dec 2014 16.5 71.4 18.5 90.0 1022.0 115.2
Jan 2015 12.6 68.2 27.2 225.0 1018.0 86.4
Feb 2015 13.2 67.6 25.2 315.0 1913.0 196.8
Mar 2015 15.4 66.1 25.6 315.0 1016.1 70.8
Apr 2015 16.6 62.3 23.3 360.0 1013.9 43.4
May 2015 21.3 66.7 18.7 225.0 1011.1 15.4
Jun 2015 24.2 66.9 17.2 225.0 1009.1 0.0
Jul 2015 27.6 65.1 15.7 225.0 1006.8 0.0
Aug 2015 29.0 64.2 15.4 225.0 1007.8 0.0
Sep 2015 27.4 65.4 17.0 180.0 1010.7 6.0
Oct 2015 23.6 67.1 19.6 225.0 1013.3 15.4
Nov 2015 19.3 60.7 19.0 45.0 1016.9 9.2
Dec 2015 14.5 63.0 21.6 45.0 1022.0 98.6
Jan 2016 12.2 65.1 25.4 270.0 1018.4 62.4
Feb 2016 15.0 69.7 20.3 315.0 1018.9 31.8
Mar 2016 16.0 65.2 25.2 315.0 1013.3 63.6
Apr 2016 19.5 65.8 20.8 360.0 1012.2 4.8
May 2016 21.3 65.4 25.4 225.0 1011.1 76.4
Jun 2016 26.0 64.6 16.3 225.0 1008.3 0.0
Jul 2016 28.6 66.9 15.8 225.0 1005.2 0.0
Aug 2016 28.7 67.6 15.7 225.0 1007.3 0.0
Sep 2016 26.4 62.4 17.6 360.0 1011.0 5.2
Oct 2016 23.2 65.1 19.8 315.0 1014.0 1.8
Nov 2016 18.8 59.5 19.9 45.0 1018.0 110.0
Dec 2016 12.9 67.5 27.3 360.0 1018.5 369.6
84

Open 12. Rainfall Prediction Using Machine Learning 3

Uploaded by

Copyright:

Available Formats

You might also like

Open 12. Rainfall Prediction Using Machine Learning 3

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Open 12. Rainfall Prediction Using Machine Learning 3

Uploaded by

Copyright:

Available Formats

ABSTRACT

Keywords: machine learning; neuro-fuzzy; neural networks; parameters; rainfall prediction

ABSTRACT .................................................................................................................................. iii

TABLE OF CONTENTS ............................................................................................................. v

LIST OF FIGURES ................................................................................................................... viii

LIST OF TABLES ....................................................................................................................... xi

LIST OF ABBREVIATIONS .................................................................................................... xii

CHAPTER 1: INTRODUCTION ................................................................................................ 1

1.1 Aim of the Study ................................................................................................................... 3

CHAPTER 2: LITERATURE REVIEW ................................................................................... 6

CHAPTER 3: RAINFALL ......................................................................................................... 10

3.1 Introduction to rainfall ........................................................................................................ 10

CHAPTER 4: MACHINE LEARNING TECHNOLOGIES .................................................. 20

4.1 Introduction to machine learning ........................................................................................ 20

CHAPTER 5: SIMULATION ................................................................................................... 28

5.1 Data processing ................................................................................................................... 28

CHAPTER 6: DISCUSSION AND RESULT........................................................................... 65

6.1 Comparing results ............................................................................................................... 65

APPENDIX A: DATABASE FOR ERBIL .................................................................................. 80

APPENDIX B: DATABASE FOR NICOSIA .............................................................................. 81

APPENDIX C: DATABASE FOR FAMAGUSTA ..................................................................... 83

Figure 3.1: Heavy and unstable clouds of conventional rainfall .................................................. 12

Figure 3.3: Cyclonic rainfall cloud formation ............................................................................. 14

Figure 4.1: Neuron scheme .......................................................................................................... 21

Figure 4.2: Structure of ANN ...................................................................................................... 22

Figure 4.3: ANFIS architecture ................................................................................................... 25

Figure 5.3: Trends in the distribution of rainfall for Erbil ........................................................... 31

Figure 5.4: Monthly average temperature for Erbil ..................................................................... 32

Figure 5.5: Trends in humidity for Erbil ...................................................................................... 32

Figure 5.6: Average wind speed for Erbil .................................................................................... 33

Figure 5.11: Trends in distribution of rainfall for Nicosia .......................................................... 37

Figure 5.12: Average temperature for Nicosia............................................................................. 37

Figure 5.13: Trends in humidity for Nicosia................................................................................ 38

Figure 5.14: Trends in wind speed for Nicosia ............................................................................ 38

Figure 5.15: Trends in average air pressure for Nicosia .............................................................. 39

Figure 5.16: Trends in wind direction for Nicosia....................................................................... 39

Figure 5.19: Trends in distribution of rainfall in Famagusta ....................................................... 42

Figure 5.20: Average temperature for Famagusta ....................................................................... 42

Figure 5.21: Trends in humidity for Famagusta .......................................................................... 43

Figure 5.22: Trends in average wind speed for Famagusta ......................................................... 43

Figure 5.23: Average wind direction for Famagusta ................................................................... 44

Figure 5.24: Trends in average air pressure for Famagusta ......................................................... 44

Figure 5.28: Snapshot of regression using NNTOOL.................................................................. 51

Figure 5.29: Proposed NARX network architecture for Erbil ..................................................... 52

Figure 5.30: Training state for NARX ......................................................................................... 52

Figure 5.32: Snapshot of regression using NNTOOL.................................................................. 54

Figure 5.33: Proposed NARX network architecture for Nicosia ................................................. 54

Figure 5.34: Training state for NARX ......................................................................................... 55

Figure 5.35: Proposed back propagation network architecture for Famagusta............................ 56

Figure 5.36: Snapshot of regression using NNTOOL.................................................................. 57

Figure 5.37: Proposed NARX network architecture .................................................................... 57

Figure 5.38: Training state for NARX ......................................................................................... 58

Figure 5.39: Proposed ANFIS architecture for Erbil ................................................................... 59

Figure 5.40: ANFIS rule viewer for Erbil.................................................................................... 60

Figure 5.42: ANFIS Rule viewer for Nicosia .............................................................................. 61

Figure 5.43: Proposed ANFIS architecture for Famagusta .......................................................... 62

Figure 5.44: ANFIS Rule Viewer for Famagusta ........................................................................ 62

Table 6.1: Obtained RMSE after training and testing .................................................................. 65

AI: Artificial intelligence